CA2370976C - Detection of nucleic acid reactions on bead arrays - Google Patents

Abstract

The present invention is directed to methods and compositions for the use of microsphere arrays to detect and quantify a number of nucleic acid reactions. The invention finds use in genotyping, i.e. the determination of the sequence of nucleic acids, particularly alterations such as nucleotide substitutions (mismatches) and single nucleotide polymorphisms (SNPs). Similarly, the invention finds use in the detection and quantification of a nucleic acid target using a variety of amplification techniques, including both signal amplification and target amplification. The methods and compositions of the invention can be used in nucleic acid sequencing reactions as well. All applications can include the use of adapter sequences to allow for universal arrays.

Description

DETECTION OF NUCLEIC ACID REACTIONS ON BEAD ARRAYS
FIELD OF THE INVENTION

The present invention is directed to methods and compositions for the use of micxosphere anays to detect and quantify a number of nucieic add reactions. The invention finds use in genotyping, I.e. the determination of the sequence of nucieic acids, parfjcuiariy alterations such as nucieofide substitutions (mismatches) and single nucieotide poiymorphisms (SNPs). Similarly, the invention finds use in the detec6on and quantfication of a nucleic acid target using a variety of ampiification techniques, including both signai ampiification and target ampiification. The methods and composiBons of the invention can be used In nucleic acid sequencing readions as well. All applkmtions can indude the use of adapter sequences to allow for universal arrays.

BACKGROUND OF THE INVENTION

The detection of specific nucieic acids Is an important tool for diagnostic medidne and molecular biology research. Gene probe assays currentiy play roles in identifying infecUous organisins such as bacteria and viruses, in probing the expression of normal and mutant genes and identifying mutant genes such as oncogenes, In typing tissue for compatibiiity preceding tissue transpiantatlon, in matching tissue or blood samples for forensic medicine, and for exploring homology among genes from different species.

Ideally, a gene probe assay should be sensitive, speciflc and easily automatabie (for a review, see .20 Nickerson, Current Opinion in Biotechnology 4:48-51 (1993)). The requirement for sensitivity (i.e. Iow detecfion limits) has been greatly alleviated by the development of the polymerase chain reaction 509,13-4 (PCR) and other amplification technologies which allow researchers to amplify exponentially a specific nucleic acid sequence before analysis (for a review, see Abramson et al., Current Opinion In Biotechnology, 4:41-47 (1993)).

Sensitivity, i.e. detectlon iimits, remain a significant obstacle in nucleic acid detec6on systems, and a variety of techniques have been developed to address this Issue. Briefly, these techniques can be classified as either target amplification or signal amplification. Target amplification invoives the amplification (f.e. replication) of the target sequence to be detected, resuiting In a significant increase in the number of target molecules. Target arriplification strategies include the polymerase chain reaction (PCR), strand displacement amplification (SDA), and nucieic.acid sequence based _10 amplificafion (NASBA).

Aitemativeiy, rather than amplify the target, alternate techniques use the target as a template to replicate a signalling probe, allowing a small number of target molecules to result In a large number of signalling probes, that then can be detected. Signal ampiification strategies include the ligase chain reaction (LCR), cycling probe technology (CPT), invasive cleavage techniques such as InvaderTM
technology, Q-Beta replicase (Q(3R) technology, and the use of'amplification probes" such as 'branched DNA' that result in multipie label probes binding to a single target sequence.

The polymerase chain reaction (PCR) Is widely used and described, and involves the use of primer extension combined with thermal cycling to amplify a target sequence; see U.S.
Patent Nos. 4,683,195 and 4,683,202, and PCR Essential Data, J. W. Wiley & sons, Ed. C.R.
Newton,1995.
In addition, there are a number of variations of PCR which also find use in the invention, including'quantitative competitive PCR" or'QC-PCR", "arbitrarily primed PCR" or "AP-PCR","immuno-PCR", "Alu-PCR",'PCR single strand conformational polymorphism' or'PCR-SSCP", allelic PCR (see Newton et al. Nucl. Acid Res.17:2503 91989); "reverse transcriptase PCR" or "RT-PCR', "biotin capture PCR", "vectorette PCR". "panhandle PCR", and'PCR
select cDNA
subtraction", among others.

Strand displacement ampiification (SDA) is generally described in Walker et al., in Molecular Methods for Vrus Detection, Academic Press, Inc., 1995, and U.S. Patent Nos. 5,455,166 and 5,130,238.
Nucleic acid sequence based amplification (NASBA) is generally described in U.S. Patent No.
5,409,818 and "Profifing from Gene-based Diagnostics", CTB Intemational Publishing Inc., N.J., 1996.
Cycling probe technology (CPT) Is a nucleic acid detection system based on signal or probe 509.13-4 ampiification rather than target ampiification, such as is done In polymerase chain reactions (PCR).
Cycling probe technology relies on a molar excess of labeled probe which contains a scissile linkage of RNA. Upon hybridization of the probe to the target, the resuiting hybrid contains a portion of RNA:DNA. This area of RNA:DNA duplex is recognized by RNAseH and the RNA is excised, resuitlng in cleavage of the probe. The probe now consists of two smaller sequences which may be released, thus leaving the target intact for repeated rounds of the readion. The unreacted probe is removed and the label is then detected. CPT is generally desctibed in U.S. Patent Nos.
5,011,769, 5,403,711, 5,660,988, and 4,876,187, and PCT pubqshed appiications WO 95105480, WO
95/1416, and WO
95/00667.

The oiigonucieotide I'iga6on assay (OLA; sometimes referred to as the ligation chain reacbon (LCR)) involve the ligation of at least two smaller probes Into a single long probe, using the target sequence as the template for the ligase. See generally U.S. Patent Nos. 5,185,243;
5,679,524 and 5,573,907;
EP 0 320 308 81; EP 0 336 731 BI; EP 0 439 182 BI; WO 90/01069; WO 89/12696;
and WO
89/09835.
InvaderTM technology is based on strudure-spedfic polymerases that cieave nucieic adds in a site-specific manner. Two probes are used: an "invader' probe and a'signaiiing' probe, that adjacently hybridize to a target sequence with a non-complementary overlap. The enzyme deaves at the overlap due to its recognition.of the `taii", and releases the "tail" with a label.
This can then be detected. The InvaderTM technology is described In U.S. Patent Nos. 5,846,717; 5,614,402;
5,719,028; 5,541,311;
and 5,843,665.

"Rolling circie ampiification" Is based on extension of a circular probe that has hybridized to.a target sequence. A polymerase is added that extends the probe sequence. As the dreuiar probe has no terminus, the polymerase repeatedly extends the circular probe resulting in concatamers of the dreuiar probe. As such, the probe is ampified. Roliing-drde ampiificaton is generally descxibed In Baner et a/. (1998) Nuc. Ackkls Res. 26:5073-5078; Barany, F. (1991) Proc. Nad. Aced Sd. USA 88:189-193;
and Lizardi et al. (1998) Nat Genet 19225-232.

"Branched DNA' signal ampiification relies on the synthesis of branched nucieic adds, containing a muitipiicity of nucieic acid "arms" that fundion to Increase the amount of label that can be put onto one probe. This technology is generally descdbed in U.S. Patent Nos. 5,681,702, 5,597,909, 5,545,730, 5,594,117, 5,591,584, 5,571,670, 5,580,731, 5,571,670; 5,591,584, 5,624,802, 5,635,352, 5,594,118, 5,359,100, 5,124,246 and 5,681,697.

Similarily, dendrimers of nucleic acids serve to vastiy increase the amount of label that can be added .
to a single molecule, using a similar idea but different compositions. This technology is as described in U.S. Patent No. 5,175,270 and Nilsen et al., J. Theor. Biol.187273 (1997).

Specificiiy, in contrast, remains a probiem-in many currentiy available gene probe assays. The extent of molecular complementariiy between probe and target defines the speciflcky of the interaction. In a practical sense, the degree of similariiy between the target and other sequences In the sample also has an impact on specificity. Variations in the concentrations of probes, of targets and of safts In the hybridization medium, In the reaction temperature, and in the length of the probe may aiter or influence the specficity of the probe/target interactlon.

It may be possible under some circumstances to distinguish targets with perfect complementarity from targets with mismatches; this Is generally very difficult using traditional technology such as filter hybridization, in situ hybridization etc., since small variations in the readion condi6ons w81 aiter the hybridization, although this may not be a problem If appropriate mismatch cantrols are provided. New experimental techniques for mismatch detection with standard probes include DNA ligation assays where single point mismatches prevent ligation and probe digestion assays in wfiich mismatches create sites for probe cleavage.

Recent focus has been on the analysis of the relationship between genetic variation and phenotype by maldng use of polymorphic DNA markers. Previous work ublized short tandem repeats (STRs) as polymorphic positionai markers; however, recent focus is on the use of single nucleotide polymorphisms (SNPs), which occur at an average frequency of more than I per ldlobase in human genomic DNA. Some SNPs, particularly those In and around coding sequences, are likely to be the direct cause of therapeu8cally relevant phenotypic variants and/or disease predisposition. There are a number of well known polymorphisms that cause clinically important phenotypes;
for example, the apoE2/3/4 variants are associated with different relative risk of Alzheimer's and other diseases (see Cordor et al., Science 261(1993). Multiplex PCR amplification of SNP lod wlth subsequent hybridization to o8gonucleotide arrays has been shown to be an 'accurate and refiable method of simultaneously genotyping at least hundreds of SNPs; see Wang et al., Sdence, 280:1077 (1998);
see also Schafer et al., Nature Biotechnology 16:33-39 (1998). The compositions of the present invention may easily be substituted for the arrays of the prior art.

There are a variety of particular techniques that are used to detect sequence, including mutations and SNPs. These include, but are not limited to, l'igation based assays, cleavage based assays (mismatch and invasive cleavage such as InvaderTM'), single base extension methods (see WO 92115712, EP 0 371 437 B1, EP 0317 074 B1; Pastinen et al., Genome Res. 7:606-614 (1997);
Syvfinen, Clinica Chimica Acta 226:225-236 (1994); and WO 91/13075), and competitive probe analysis (e.g.
competidve sequencing by hybridization; see below).

'50~13-4 In addition, DNA sequencing is a crucial technology in biology today, as the rapid sequencing of genomes, inciuding the human genome, is both a signiflcant goal and a signiflcant hurdle. Thus there is a significant need for robust, high-throughput methods. Traditionally, the most common method of DNA sequencing has been based on polyacrylamide gei fractionatjon to resolve a popuiation of chain-terminated fragments (Sanger et al., Proc. NaU. Acad. Sci. USA 74:5463 (1977);
Maxam & Gilberi:).
The popuiation of fragments, terminated at each position in the DNA sequence, can be generated in a number of ways. Typically, DNA polymerase Is used to incorporate dideoxynucieofdes that serve as chain terminators.

Several aitemative methods have been developed to Increase the speed and ease of DNA
sequencing. For example, sequencing by hybridization has been descn'bed (Drmanac et al., Genomics 4:114 (1989); Koster et al., Nature Biotechnology 14:1123 (1996);
U.S. Patent Nos.

5,525,464; 5,202,231 and 5,695,940, among others). Similarly, sequencing by synthesis Is an aitemat'rve to gel-based sequencing. These methods add and read only one base (or at most a few bases, typically of the same type) prior to poiymerization of the next base.
This can be referred to as "tlme resoived" sequencing, to contrast from "gel-resoived" sequencing.
Sequendng by synthesis has been described in U. S. Patent No 4,971,903 and Hyman, Anal. Biochem.174:423 (1988); Rosenthal, tntemational Patent Appiication Pubiication 761107 (1989); Metzker et al., Nucl. Acids Res. 22:4259 (1994); Jones, Biotechniques 22:938 (1997); Ronaghl et al., Anal. Biochem.
242:84 (1996), Nyren et al., Anal. Biochem.151:504 (1985). Detection of ATP sulfurylase adivity is described In Karamohamed and Nyren, Anal. Blochem. 271:81 (1999). Sequencing using reversible chain terminating nucieotides is described in U.S. Patent Nos. 5,902,723 and 5,547,839, and Canard and Arzumanov, Gene 11:1 (1994), and Dyatkina and Arzumanov, Nucleic Acids Symp Ser 18:117 (1987).
Reversible chain termination with DNA ligase is described in U.S. Patent 5,403,708. Time resolved sequencing is described in Johnson et al., Anal. Biochem. 136:192 (1984).
Single molecule analysis is described in U.S. Patent No. 5,795,782 and Elgen and Rigler, Proc. Nati Acad Sd USA 91(13):5740 (1994).

One promising sequencing by synthesis method is based on the detection of the pyrophosphate (PPi) released during the DNA polymerase reaction. As nucieotriphosphates are added to a growing nucleic acid chain, they release PPI. This release can be quantitativeiy measured by the conversion of PPi to ATP by the enzyme suifuryiase, and the subsequent produdion of visible light by firefly luciferase.
Several assay systems have been described that capitalize on this mechanism.
See for example WO 93/23564, WO 98/28440 and WO 98/13523. A preferred method is described in Ronaghi et al., Science 281:363 (1998). In this method, the four deoxynucleotides (dATP, dGTP, dCTP and dTTP; collectiveiy dNTPs) are added stepwise to a pargal duplex comprising a sequencing primer hybridized to a single stranded DNA
template and incubated with DNA polymerase, ATP sulfurylase, luciferase, and optionally a nucieotide-degrading enzyme such ti:qV9 13-4 as apyrase. A dNTP is only Incorporated into the growing DNA strand if it is complementary to the base in the template strand. The synthesis of DNA Is accompanied by the release of PPI equal In moiarity to the incorporated dNTP. The PPI is converted to ATP and the light generated by the ludferase Is directly proportionai to the amount of ATP. In some cases the unincorporated dNTPs and the produced ATP are degraded between each cycle by the nucieotide degrading enzyme.

In some cases the DNA template Is associated with a solid supporL To this end, there are a wide variety of known methods of attaching DNAs to solid supports. Recent work has focused on the attachment of binding ligands, including nucleic acid probes, to microspheres that are randomly d'istributed. on a surface, including a fiber optic bundle, to form high density arrays. See for example PCTs U898/21193, PCT US99/14387 and PCT US98/05025; W098/50782.

An additional technique utiiizes sequencing by hybridization. For example, sequencing by hybridization has been described (Drmanac et al., Genomics 4:114 (1989); U.S.
Patent Nos.
5,525,464; 5,202,231 and 5,695,940, among others.

In addition, sequencing using mass spectrometry techniques has been described;
see Koster et al., Nature Biotechnology 14:1123 (1996).

Finally, the use of adapter-type sequences that allow the use of universal arrays has been described In limited contexts; see for example Chee et al., Nucl. Add Res. 19:3301 (1991);
Shoemaker et al., Nature Genetics 14:450 (1998); Barany, F. (1991) Proc. Nab. Acad. Sci. USA
88:189-193; EP 0 799 897 Al; WO 97/31256.

PCTs US98/21193, PCT US99/14387 and PCT US98/05025; W098/50782; and describe novel compositions utilizing substrates with microsphere arrays, which .25 allow for novel detection methods of nucleic acid hybridization.

Accordingly, it Is an object of the present Invention to provide deteciton and quantfication methods for a variety of nucieic acid reactions, inciuding genotyping, ampificatlon reactions and sequendng reactions, utiiizing microsphere arrays.

SUMMARY OF THE INVENTION

In accordance with the above objects, the present invention provides methods of determining the identity of a nucleotide at a detection posfion in a target sequence. The methods comprise providing a hybridization complex comprising the target sequence and a capture probe covalently attached to a microsphere on a surface of a substrate. The methods comprise determining the nucleotide at the detection posifion. The hybridization complex can comprise the capture probe, a capture extender probe, and the target sequence. In addition, the target sequence may comprise exogeneous adapter sequences.

In an additional aspect, the method comprises contacfing the microspheres with a plurality of detection probes each comprising a unique nucleotide at the readout position and a unique detectable label.
The signal from at least one of the detectable labels is detected to identify the nucleotide at the detection position.

In an additional aspect, the detecton probe does not contain detection label, but rather is identified based on its characterisfic mass, for example via mass spectrometry. In addition, the detection probe comprises a unique label that is detected based on its characteristic mass.

In a further aspect, the invention provides methods wherein the target sequence comprises a first target domain directly 5' adjacent to the detecton position. The hybridization complex comprises the target sequence, a capture probe and an extension primer hybridized to the first target domain of the target sequence. The determinafion step comprises contacting the microspheres with a polymerase enzyme, and a plurality of NTPs each comprising a covalently attached detectable label, under conditions whereby if one of the NTPs basepairs with the base at the detecton position, the extension primer is extended by the enzyme to incorporate the label. As is known to those in the art, dNTPs and ddNTPs are the preferred substrates for DNA polymerases. NTPs are the preferred substrates for RNA polymerases. The base at the detection position is then identified.

In an additional aspect, the invention provides methods wherein the target sequence comprises a first target domain directly 5' adjacent to the detecton position, wherein the capture probe serves as an extension primer and is hybridized to the first target domain of the target sequence. The determination step comprises contacfing the microspheres with a polymerase enzyme, and a plurality of NTPs each comprising a covalently attached detectable label, under conditions whereby if one of the NTPs basepairs with the base at the detection position, the extension primer is extended by the enzyme to incorporate the label. The base at the detecton position is thus identified.

In a further aspect, the invention provides methods wherein the target sequence comprises (5' to 3'), a first target domain comprising an overlap domain comprising at least a nucleotide in the detection position and a second target domain configuous with the detecton position. The hybridizafion complex comprises a first probe hybridized to the first target domain, and a second probe hybridized to the second target domain. The second probe comprises a detection sequence that does not hybridize with the target sequence, and a detectable label. If the second probe comprises a base that is perfectly complementary to the detection position a cleavage structure is formed. The method further comprises contacfing the hybridization complex with a cleavage enzyme that will cleave the detecfion sequence from the signalling probe and then forming an assay complex vrith the detecfion sequence, -5 a capture probe covalently attached to a microsphere on a surface of a substrate, and at least one label. The base at the detection position is thus identified.

In an additional aspect, the invenfion provides methods of determining the identification of a nucleotide at a detection position in a target sequence comprising a first target domain comprising the detection position and a second target domain adjacent to the detection position. The method comprises hybridizing a first ligation probe to the first target domain, and hybridizing a second ligation probe to the second target domain. If the second ligafion probe comprises a base that is perfectly complementary to the detecfion position a ligation structure is formed. A ligation enzyme is provided that will ligate the first and the second ligafion probes to form a ligated probe. An assay complex is formed vvith the ligated probe, a capture probe covalently attached to a microsphere on a surface of a substrate, and at least one label. The base at the detection posifion is thus identified.

In a further aspect, the present invenfion provides methods of sequencing a plurality of target nucleic acids. The methods comprise providing a plurality of hybridization complexes each comprising a target sequence and a sequencing primer that hybridizes to the first domain of the target sequence, the hybridization complexes are attached to a surface of a substrate. The methods comprise extending each of the primers by the addition of a first nucleotide to the first detection position using an enzyme to form an extended primer. The methods comprise detecting the release of pyrophosphate (PPi) to determine the type of the first nucleotide added onto the primers. In one aspect the hybridizafion complexes are attached to microspheres distributed on the surface. In an additional aspect the sequencing primers are attached to the surface. The hybridization complexes comprise the target sequence, the sequencing primer and a capture probe covalently attached to the surface. The hybridization complexes also comprise an adapter probe.

In an additional aspect, the method comprises extending the extended primer by the addi6on of a second nucleotide to the second detection position using an enzyme and detecting the release of pyrophosphate to determine the type of second nucleotide added onto the primers. In an additional aspect, the pyrophosphate is detected by contacting the pyrophosphate with a second enzyme that converts pyrophosphate into ATP, and detecting the ATP using a third enzyme.
In one aspect, the second enzyme is sulfurylase and/or the third enzyme is luciferase.

In an additional aspect, the invention provides methods of sequencing a target nucleic acid comprising a first domain and an adjacent second domain, the second domain comprising a plurality of target positions. The method comprises providing a hybridization complex comprising the target sequence and.a capture probe covalently attached to microspheres on a surface of a substrate and determining the identity of a plurality of bases at the target positions, The hybridization complex comprises the capture probe, an adapter probe, and the target sequence. In one aspectthe sequencing primer is the capture probe.

In an additional aspect of the invention, the determining comprises providing a sequencing primer hybridized to the second domain, extending the primer by the addition of first nucleotide to the first detection position using a first enzyme to form an extended primer, detecfing the release of pyrophosphate to determine the type of the first nucleotide added onto the primer, extending the primer by the addi6ion of a second nucleotide to the second detection position using the enzyme, and detecting the release of pyrophosphate to determine the'type of the second nucleotide added onto the primer. In an additional aspect pyrophosphate is detected by contac6ng the pyrophosphate with the second enzyme that converts pyrophosphate into ATP, and detec6ng the ATP using a third enzyme.
In one aspect the second enzyme is sulfurylase and/or the third enzyme is luciferase.

In an additional aspect of the method for sequencing, the determining comprises providing a sequencing primer hybridized to the second domain, extending the primer by the addifion of a first protected nucleotide using a first enzyme to form an extended primer, determining the idenfification of the first protected nucleotide, removing the protecfion group, adding a second protected nucleotide using the enzyme, and determining the identificafion of the second protected nuclea6de.

According to a preferred embodiment of the invention, there is provided a method of sequencing a plurality of target nucleic acids each comprising a first domain and an adjacent second domain, said second domain comprising a plurality of detection positions, said method comprising: a) providing a plurality of hybridization complexes each comprising a target sequence and a sequencing primer that hybridizes to the first domain of said target sequence, wherein said hybridization complexes are attached to sites on an array, said array comprising at least a first substrate with a surface comprising individual sites; b) extending each of said primers by the addition of a first nucleotide to the first detection position using a first enzyme to form an extended primer; and c) detecting the release of pyrophosphate (PPi) to determine the type of said first nucleotide added onto said primers, wherein said release of PPi is detected by secondary enzymes, said secondary enzymes being attached to sites on said array.

in an additional aspect the invention provides a kit for nucleic acid sequencing comprising a composition comprising a substrate with a surface comprising discrete sites and a population of microspheres distributed nn the sites, wherein the microspheres comprise capture probes. The kit also comprises an extension enzyme and dNTPs. The kit also comprises a second enzyme for the conversion of pyrophosphate to ATP and a third enzyme for the detecfion of ATP. In one aspect the dNTPs are labeled. In addition each dNTP comprises a different label.

According to a preferred embodiment of the invention, there is provided a kit for nucleic acid sequencing comprising: a) a composition comprising: i) a substrate with a surface comprising discrete sites, and ii) a population of microspheres distributed on said sites; wherein said microspheres comprise capture probes for hybridizing to target nucleic acid sequences; b) an extension enzyme for .5 enzymatically extending an oligonucleotide chain; c) dNTPs; and d) secondary enzymes for detecting the release of pyrophosphate (PPi), said secondary enzymes being attached to microspheres.

In a further aspect, the present invenfion provides methods of detec4ng a target nucleic acid sequence comprising attaching a first adapter nucleic acid to a first target nucleic acid sequence to form a modified first target nucleic acid sequence, and contacfing the modified first target nucleic acid sequence vvith an array as outlined herein. The presence of the modified first target nucleic acid sequence is then detected.

In an addi6onal aspect, the methods further comprise attaching a second adapter nucleic acid to a second target nucleic acid sequence to form a modified second target nucleic acid sequence and contacfing the modified second target nucleic acid sequence with the array.

In a further aspect, the invention provides methods of detecting a target nucleic acid sequence 9a comprising hybridizing a first primer to a first portion of a target sequence, wherein the first primer further comprises an adapter sequence and hybridizing a second primer to a second por6on of the target sequence. The first and second primers are ligated together to form a modified primer, and the adapter sequence of the modified primer is contacted with an array of the invention, to allow detection of the presence of the modified primer.

In an additional embodiment, the present invention provides a method for detecting a first target nucleic acid sequence. In one aspect the method comprises hybridizing at least a first primer nucleic acid to the first target sequence to form a first hybridization complex, contacting the first hybridization complex with a first enzyme to form a modified first primer nucleic acid, disassociating the first hybridization complex, contacting the modified first primer nucleic acid with an array comprising a substrate with a surface comprising discrete sites and a population of microspheres comprising at least a first subpopulation comprising a first capture probe such that the first capture probe and the modified primer form an assay complex, wherein the microspheres are distributed on the surface, and detecting the presence of the modified primer nucleic acid.

In addifion the method further comprises hybridizing at least a second primer nucleic acid to a second target sequence that is substantially complementary to the first target sequence to form a second hybridization complex, contacting the second hybridization complex with the first enzyme to form modified second primer nucleic acid, disassociating the second hybridization complex and forming a second assay complex comprising the modified second primer nucleic acid and a second capture probe on a second subpopulation.

In an addifional aspect of the invenfion the primer forms a circular probe following hybridization with the target nucleic acid to form a first hybridization complex and contacting the first hybridization complex with a first enzyme comprising a ligase such that the oligonucleotide ligation assay (OLA) occurs. This is followed by adding the second enzyme, a polymerase, such that the circular probe is amplified in a rolling circle amplification (RCA) assay.

In an additional aspect of the invention, the first enzyme comprises a DNA
polymerase and the modification is an extension of the primer such that the polymerase chain reaction (PCR) occurs. In an addifional aspect of the invention the first enzyme comprises a ligase and the modificafion comprises a ligation of the first primer which hybridizes to a first domain of the first target sequence, to a third primer which hybridizes to a second adjacent domain of the first target sequence such that the ligase chain reaction (LCR) occurs.

In an addifional aspect of the invenfion, the first primer comprises a first probe sequence, a first scissile linkage and a second probe sequence, wherein the first enzyme will cleave the scissile linkage resulting in the separation of the first and second probe sequences and the disassociafion of the first hybridization complex, leaving the first target sequence intact such that the cycling probe technology (CPT) reaction occurs.

In addition, wherein the first enzyme is a polymerase that extends the first primer and the modified first primer comprises a first newly synthesized strand, the method further comprises the addition of a second enzyme comprising a nicking enzyme that nicks the extended first primer leaving the first target sequence intact, and extending from the nick using the polymerase, and thereby displacing the first newly synthesized strand and generating a second newly synthesized strand such that strand displacement amplification (SBA) occurs.

In addition, wherein the first target sequence is an RNA target sequence, the first primer nucleic acid is a DNA primer comprising an RNA polymerase promoter, the first enzyme is a reverse-transcriptase that extends the first primer to form a first newly synthesized DNA strand, the method further comprises the addition of a second enzyme comprising an RNA degrading enzyme that degrades the first target sequence, the addition of a third primer that hybridizes to the first newly synthesized DNA
strand, the addition of a third enzyme comprising a DNA polymerase that extends the third primer to form a second newly synthesized DNA strand, to form a newly synthesized DNA
hybrid, the addition of a fourth enzyme comprising an RNA polymerase that recognizes the RNA
polymerase promoter and generates at least one newly synthesized RNA strand from the DNA hybrid, such that nucleic acid sequence-based amplification (NASBA) occurs.

In addition, wherein the first primer is an invader primer, the method further comprises hybridizing a signalling primer to the target sequence, the enzyme comprises a structure-specific cleaving enzyme and the modification comprises a cleavage of said signalling primer, such that the invasive cleavage reaction occurs.

An additional aspect of the invention is a method for detecting a target nucleic acid sequence comprising hybridizing a first primer to a first target sequence to form a first hybridization complex, contacfing the first hybridization complex with a first enzyme to extend the first primer to form a first newly synthesized strand and form a nucleic acid hybrid that comprises an RNA
polymerase promoter, contacfing the hybrid with an RNA polymerase that recognizes the RNA
polymerase promoter and generates at least one newly synthesized RNA strand, contacting the newly synthesized RNA strand with an array comprising a substrate with a surface comprising discrete sites and a population of microspheres comprising at least a first subpopulation comprising a first capture probe; such that the first capture probe and the modified primer form an assay complex; wherein the microspheres are distributed on the surface and detecting the presence of the newly synthesized RNA strand.

In addition, when the target nucleic acid sequence is an RNA sequence, and prior to hybridizing a first primer to a first target sequence to form a first hybridization complex, method comprises hybridizing a second primer comprising an RNA polymerase promoter sequence to the RNA
sequence to form a second hybridization complex, contacting the second hybridization complex with a second enzyme to extend the second primer to form a second newly synthesized strand and form a nucleic acid hybrid;
and degrading the RNA sequence to leave the second newly synthesized strand as the first target sequence. In one aspect of the invention the degrading is done by the addition of an RNA degrading enzyme. In an additional aspect of the invention the degrading is done by RNA
degrading activity of reverse transcriptase.

In addition, when the target nucleic acid sequence is a DNA sequence, and prior to hybridizing a first primer to a first target sequence to form a first hybridization complex, the method comprises hybridizing a second primer comprising an RNA polymerase promoter sequence to the DNA sequence to form a second hybridization complex, contacfing the second hybridization complex with a second enzyme to extend the second primer to form a second newly synthesized strand and form a nucleic acid hybrid, and denaturing the nucleic acid hybrid such that the second newly synthesized strand is the first target sequence.

An additional aspect fo the invention is a kit for the detection of a first target nucleic acid sequence.
The kit comprises at least a first nucleic acid primer substandally complementary to at least a first domain of the target sequence, at least a first enzyme that will modify the first nucleic acid primer, and an array comprising a substrate with a surface comprising discrete sites, and a population of microspheres comprising at least a first and a second subpopulation, wherein each subpopulation comprises a bioactive agent, wherein the microspheres are distributed on the surface.

In an additional aspect of the invenfion, is a kit for the detection of a PCR
reaction wherein the first enzyme is a thermostable DNA polymerase.

In an additional aspect of the invention, is a kit for the detection of a LCR
reaction wherein the first enzyme is a ligase and the kit comprises a first nucleic acid primer substantially complementary to a first domain of the first target sequence and a third nucleic acid primer substanfially complementary to a second adjacent domain of the first target sequence.

In an additional aspect of the invention, is a kit for the detection of a strand displacement amplification (SDA) reacfion wherein the first enzyme is a polymerase and the kit further comprises a nicking enzyme.

In an additional aspect of the invention, is a kit for the detection of a NASBA reacfion wherein the first enzyme is a reverse transcriptase, and the kit comprises a second enzyme comprising an RNA
degrading enzyme, a third primer, a third enzyme comprising a DNA polymerase and a fourth enzyme comprising an RNA polymerase.

In an additional aspect of the invention, is a kit for the detection of an invasive cleavage reaction wherein the first enzyme is a structure-specific cleaving enzyme, and the kit comprises a signaling primer.

BRIEF DESCRIPTION OF THE DRAWINGS

Figures 1 A, 1 B and 1 C depict three different embodiments for attaching a target sequence to an array.
The solid support 5 has microsphere 10 with capture probe 20 linked via a linker 15. Figure 1A
depicts direct attachment; the capture probe 20 hybridizes to a first porbon of the target sequence 25.
Figure 1 B depicts the use of a capture extender probe 30 that has a first porbon that hybridizes to the capture probe 20 and a second por6on that hybridizes to a first domain of the target sequence 25.
Figure 1 C shows the use of an adapter sequence 35, that has been added to the target sequence, for example during an amplification reaction as outlined herein.

Figures 2A and 2B depict two preferred embodiments of SBE amplification.
Figure 2A shows extension primer 40 hybridized to the target sequence 25. Upon addifion of the extension enzyme and labelled nucleotides, the extension primer is modified to form a labelled primer 41. The reaction can be repeated and then the labelled primer is added to the array as above.
Figure 2B depicts the same reaction but using adapter sequences.

Figures 3A and 3B depict two preferred embodiments of OLA amplificafion.
Figure 3A depicts a first ligation probe 45 and a second ligation probe 50 with a label 55. Upon addition of the ligase, the probes are ligated. The reacfion can be repeated and then the ligated primer is added to the array as above. Figure 3B depicts the same reaction but using adapter sequences.

Figure 4 depicts a preferred embodiment of the invasive cleavage reaction. In this embodiment, the signaling probe 65 comprises two portions, a detection sequence 67 and a signaling portion 66. The signaling porbon can serve as an adapter sequence. In addition, the signaling portion generally comprises the label 55, although as will be appreciated by those in the art, the label may be on the detection sequence as well. In addition, for optional removal of the uncleaved probes, a capture tag 60 may also be used. Upon addition of the enzyme, the structure is cleaved, releasing the signaling por6on 66. The reaction can be repeated and then the signaling pordon is added to the array as above.

Figures 5A and 5B depict two preferred embodiments of CPT amplification. A CPT
primer 70 comprising a label 55, a first probe sequence 71 and a second probe sequence 73, separated by a scissile linkage 72, and optionally comprising a capture tag 60, is hybridized to the target sequence 25.
Upon addition of the enzyme, the scissile linkage is cleaved. The reaction can be repeated and then the probe sequence comprising the label is added to the array as above. Figure 5B depicts the same reaction but using adapter sequences.

Figure 6 depicts OLA/RCA amplification using a single "padlock probe" 57. The padlock probe is hybridized with a target sequence 25. When the probe 57 is complementary to the target sequence 26, ligation of the probe termini occurs forming a circular probe 28. When the probe 57 is not .5 complementary to the target sequence 27, iigation does not occur. Addition of polymerase and nucleotides to the circular probe results amplification of the probe 58.
Cleavage of the amplified probe 58 yields fragments 59 that hybridize with an identifier probe 21 immobilized on a microsphere 10.
Figure 7 depicts an alternative method of OLA/RCA. An immobifized first OLA
primer 45 is hybridized with a target sequence 25 and a second OLA primer 50. Following the addition of ligase, the first and second OLA primers are ligated to form a ligated oligonucleotide 56. Following denaturafion to remove the target nucleic acid, the immobilized ligated oligonucleotide is distributed on an array. An RCA probe 57 and polymerase are added to the array resulting in amplification of the circular RCA
probe 58.

Figures 8A, 8B, BC, 8D and 8E schematically depict the use of readout probes for genotyping. Figure 8A shows a"sandwich' format. Substrate 5 has a discrete site with a microsphere 10 comprising a capture probe 20 attached via a linker 15. The target sequence 25 has a first domain that hybridizes to the capture probe 20 and a second domain comprising a detection position 30 that hybridizes to a readout probe 40 with readout posi6on 35. As will be appreciated by those in the art, Figure 8A
depicts a single detection posi6on; however, depending on the system, a plurality of different probes .20 can hybridize to different target domains; hence n is an integer of 1 or greater. Figure 8B depicts the use of a capture probe 20 that also serves as a readout probe. Figure 8C
depicts the use of an adapter probe 100 that binds to both the capture probe 20 and the target sequence 25. As will be appreciated by those in the art, the figure depicts that the capture probe 20 and target sequence 25 bind adjacently and as such may be ligated; however, as wiil be appreciated by those in the art, there may be a "gap" of one or more nucleotides. Figure 8D depicts a solution based assay. Two readout probes 40, each with a different readout position (35 and 36) and different labels (45 and 46) are added to target sequence 25 with detection position 35, to form a hybridization complex with the match probe. "This is added to the array; Figure 8D depicts the use of a capture probe 20 that directly hybridizes to a first domain of the target sequence, although other attachments may be done. Figure BE depicts the direct attachment of the target sequence to the array.

Figures 9A, 9B, 9C and 9D depict preferred embodiments for SBE genotyping.
Figure 9A
depicts a"sandwidh" assay, in which substrate 5 has a discrete site with a microsphere 10 comprising a capture probe 20 attached via a linker 15. The target sequence 25 has a first domain that hybridizes 50,913-4 to the capture probe 20 and a second domain comprising a detection position 30 that hybridizes to an extension primer 50. As will be appreciated by those in the art, Figure 9A
depicts a single detecfion position; however, depending on the system, a piurality of different primers can hybridize to different target domains; hence n is an integer of I or greater. In addfion, the first domain of the target sequence may be an adapter sequence. Figure 9B depicts the use of a capture probe 20 that also serves as an extension primer. Figure 9C depicts the soiution reactlon. Figure 9D depicts the use of a capture extender probe 100, that has a first domain that will hybridize to the capture probe 20 and a second domain that wiii hybridize to a first domain of the target sequence 25.

Figures 10A,10B,10C,10D and 10E depict some of the OLA genotyping embodiments of the reac6on. Figure 10A depicts the soiution reaction, wherein the target sequence 25 with a detectlon position 30 hybridizes to the first i'igation probe 75 with readout position 35 and second probe 76 with a detectable label 45. As will be appreciated by those In the art, the second ligation probe could also contain the readout positlon. The addition of a iigase forms a ligated probe 80, that can then be added to the array with a capture probe 20. Figure 10B depicts an "on bead' assay, wherein the capture probe 20 serves as the first 1iga5on probe. Figure 10C depicts a sandwich assay, using a capture probe 20 that hybridizes to a first portion of the target sequence 25 (which may be an endogeneous sequence or an exogeneous adapter sequence) and iigation probes 75 and 76 that hybridize fio a second portion of the target sequence comprising the detection position 30.
Figure 10D depects the use of a capture extender probe 100. Figure IOE depicts a soiution based assay with the use of an adapter sequence 110.

Figures 11A,11B and 11C depictthe SPOLA reaction forgenotyping. In Figure 11A, two (igation probes are hybridized to a target sequence. As wili be appreciated by those in the art, this system requires that the two ligation probes be attached at different ends, i.e. one at the 5' end and one at the 3' end. One of the ligation probes is attached via a cleavable linker. Upon formatlon of the assay complex and the addition of a ligase, the two probes will efficiently covalently couple the two iigation probes if perfect compiementarity at the junction exists. In Figure 118, the resuiting ligation difference between correctiy matched probes and imperfect probes is shown. Figure 11 C
shows that subsequent cleavage of the cleavable linker produces a reactive group, In this case an amine, that may be subsequentiy labeled as outlined herein. Alternatively, cleavage may leave an upstream oligo with a detectable label. If not ligated, this labeled oligo can be washed away.

Figures 12A and 12B depict two cleavage reactions for genotyping. Figure 12A
depicts a loss of signal assay, wherein a label 45 is cleaved off due to the discrimination of the cleavage enzyme such as a restriction endonuclease or resolvase type enzyme to allow single base mismatch discrimination.
Figure 12B depicts the use of a quencher 46.

Figure 13A, 13B, 13C, 13D, 13E and 13F depict the use of invasive cleavage to determine the identity of the nucleofiide at the detection position. Figures 13A and 13B depict a loss of signal assay. Figure 13A depicts the invader probe 55 with readout position 35 hybridized to the target sequence 25 which is attached via a capture probe 20 to the surface. The signal probe 60 with readout position 35, detectable label 45 and detection sequence 65 also binds to the target sequence 25; the two probes form a cleavage structure. If the two readout positions 35are capable of basepairing to the detec6on position 30 the addition of a structure-specific cleavage enzyme releases the detection sequence 65 and consequently the label 45, leading to a loss of signal. Figure 13B is the same, except that the capture probe 20 also serves as the invader probe. Figure 13C depicts a solution reaction, wherein the signalling probe can comprise a capture tag 70 to facilitate the removal of uncleaved signal probes. The addition of the cleaved signal probe (e.g. the detection sequence 65) with its associated label 45 results in detection. Figure 13D depicts a soiution based assay using a label probe 120.
Figure 13E depicts a preferred embodiment of an invasive cleavage reaction that utilizes a fluorophore-quencher reaction. Figure 13E has the 3' end of the signal probe 60 is attached to the bead 10 and comprises a label 45 and a quencher 46. Upon formation of the assay complex and subsequent cleavage, the quencher 46 is removed, leaving the fluorophore 45.

Figures 14A, 14B, 14C and 14D depict genotyping assays based on the novel combination of competitive hybridization and extension. Figures 14A, 14B and 14C depict solufion based assays.
After hybridization of the extension probe 50 with a match base at the readout position 35, an extension enzyme and dNTP is added, wherein the dNTP comprises a blocking moiety (to facilitate removal of unextended primers) or a hapten to allow purification of extended primer, i.e. biotin, DNP, fluorescein, etc. Figure 14B depicts the same reaction with the use of an adapter sequence 90; in this embodiment, the same adapter sequence 90 may be used for each readout probe for an allele.
Figure 14C depicts the use of different adapter sequences 90 for each readout probe; in this embodiment, unreacted primers need not be removed, although they may be.
Figure 14D depicts a solid phase reacfion, wherein the dNTP added in the position adjacent to the readout position 35 is labeled.

Figures 15A and 15B depict genotyping assays based on the novel combination of invasive cleavage and ligation reactions. Figure 15A is a solution reaction, with the signalling probe 60 comprising a detection sequence 65 with a detectable label 45. After hybridization with the target sequence 25 and cleavage, the free detection sequence can bind to an array (depicted herein as a bead array, although any nucleic acid array can be used), using a capture probe 20 and a template target sequence 26 for the ligation reaction. In the absence of ligation, the signalling probe is washed away. Figure 15B

depicts a solid phase assay. In this embodiment, the 5' end of the signalling probe is attached to the array (again, depicted herein as a bead array, although any nucleic acid array can be used), and a blocking moiety is used at the 3' end. After cleavage, a free 3' end is generated, that can then be used for ligafion using a template target 26. As will be appreciated by those in the art, the orientation of this may be switched, such that the 3' end of the signalling probe 60 is attached, and a free 5' end is generated for the ligation reaction.

Figures 16A and 16B depict genotyping assays based on the novel combination of invasive cleavage and extension reactions. Figure 16A depicts an initial solution based assay, using a signalling probe with a blocked 3' end. After cleavage, the detection sequence can be added to an array and a template target added, followed by extension to add a detectable label.
Alternatively, the extension can also happen in solution, using a template target 26, followed by additon of the extended probe to the array. Figure 16B depicts the solid phase reac6on; as above, either the 3' or the 5' end can be attached. By using a blocking moiety 47, only the newly cleaved ends may be extended.

Figures 17A, 17B and 17C depict three configurations of the combination of ligation and extension ("Genetic Bit" analysis) for genotyping. Figure 17A depicts a reaction wherein the capture probe 20 and the extension probe serve as two ligation probes, and hybridize adjacently to the target sequence, such that an additional ligation step may be done. A labeled nucleofide is added at the readout position. Figure 17B depicts a preferred embodiment, wherein the ligation probes (one of which is the capture probe 20) are separated by the detection position 30. The addition of a labeled dNTP, extension enzyme and ligase thus serve to detect the readout position. Figure 17C depicts the solution phase assay. As will be appreciated by those in the art, an extra level of specificity is added if the capture probe 20 spans the ligated probe 80.

DETAILED DESCRIPTION OF THE INVENTION

The present invention is directed to the detecfion and quantificafion of a variety of nucleic acid reactions, parficularly using microsphere arrays. In particular, the invention relates to the detection of amplification, genotyping, and sequencing reactions. In addition, the invention can be utilized with adapter sequences to create universal arrays.

Accordingly, the present invention provides compositions and methods for detecting and/or quantifying the products of nucleic acid reacfions, such as target nucleic acid sequences, in a sample. As will be appreciated by those in the art, the sample solution may comprise any number of things, including, but not limited to, bodily fluids (including, but not limited to, blood, urine, serum, lymph, saliva, anal and vaginal secretions, perspirafion and semen, of virtually any organism, with mammalian samples being preferred and human samples being particularly preferred); environmental samples (including, but not limited to, air, agricultural, water and soil samples); biological warfare agent samples; research samples; purified samples, such as purfied genomic DNA, RNA, proteins, etc.;
raw samples (bacteria, virus, genomic DNA, etc.; As will be appreciated by those in the art, virtually any experimental manipulation may have been done on the sample.

The present invention provides composi6ons and methods for detec6ng the presence or absence of target nucleic acid sequences in a sample. By "nucleic add" or "oiigonucleotide" or grammatical equivalents herein means at least two nucleotides covalently linked together.
A nucleic acid of the present invention will generally contain phosphodiester bonds, aithough In some cases, as outlined below, nucleic acid analogs are included that may have altemate backbones, comprising, for example, phosphoramide (Beaucage et al., Tetrahedron 49(10):1925 (1993) and references therein;
Letsinger, J. Org. Chem. 35:3800 (1970); Sprinzl et al., Eur. J. Biochem.
81:579 (1977); Letsinger et al., Nucl. Acids Res. 14:3487 (1986); Sawal et al, Chem. Lett. 805 (1984), Letsinger et al., J. Am.
Chem. Soc. 110:4470 (1988); and Pauwels et al., Chemica Scripta 26:141 91986)), phosphorothloate (Mag et al., Nucleic Acids Res. 19:1437 (1991); and U.S. Patent No.
5,644,048), phosphorodithioate (Briu et al., J. Am. Chem. Soc.1112321 (1989), O-methylphophoroamidite linkages (see Eckstein, Oligonucleotides and Analogues: A Pracfical Approach, Oxford University Press), and peptlde nucieic acid backbones and linkages (see Egholm, J. Am. Chem. Soc.114:1895 (1992);
Meier et al., Chem.
Int. Ed. Engl. 31:1008 (1992); Nielsen, Nature, 365:566 (1993); Carlsson et al., Nature 380207 (1996).
Other analog nucleic acids include those with positive backbones (Denpcy et al., Proc. Natl. Acad. Sci. USA 92:6097 (1995); non-ionic backbones (U.S.
Patent Nos. 5,386,023, 5,637,684, 5,602,240, 5,216,141 and 4,469,863;
Kiedrowshi et al., Angew.
Chem. Intl. Ed. English 30:423 (1991); Letsinger et al., J. Am. Chem.
Soc.110:4470 (1988); Letsinger et al., Nucleoside & Nucieotide 13:1597 (1994); Chapters 2 and 3, ASC
Symposium Series 580, "Carbohydrate Modifications In Antisense Research", Ed. Y.S. Sanghui and P.
Dan Cook; Mesmaeker et al., Bioorganic & Medicinal Chem. LetL 4:395 (1994); Jeffs et al., J.
Biomolecular NMR 34:17 (1994); Tetrahedron Lett. 37:743 (1996)) and non-ribose backbones, including those described in U.S.
Patent Nos. 5,235,033 and 5,034,506, and Chapters 6 and 7, ASC Symposium Series 580, "Carbohydrate Modifications in Antisense Research", Ed. Y.S. Sanghul and P.
Dan Cook. Nucleic acids containing one or more carbocyclic sugars are also included within the definition of nucleic acids (see Jenkins et al., Chem. Soc. Rev. (1995) pp169-176). Several nucieic acid analogs are described in Rawis, C & E News June 2, 1997 page 35. These modifications of the ribose-phosphate backbone may be done to facilitate the addition of labels, or to increase the stability and half-life of such molecules in physiological environments.

As wili be appreciated by those In the art, all of these nucieic acid analogs may find use in the present invention. In addition, mbctures of naturally occurring nucleic adds and analogs can be made.
Alternatively, mbctures of different nucleic acid analogs, and mixtures of naturally occuring nucleic acids and analogs may be made.

Par6cularly preferred are peptide nucleic acids (PNA) which includes peptide nucleic acid analogs.
These backbones are substantially non-ionic under neutral conditions, in contrast to the highly charged phosphodiester backbone of naturally occurring nucleic acids. This results in two advantages. First, the PNA backbone exhibits improved hybridization kinetics. PNAs have larger changes in the mel6ng temperature (Tm) for mismatched versus perfectly matched basepairs. DNA and RNA typically exhibit a 2-4 C drop in Tm for an internal mismatch. With the non-ionic PNA backbone, the drop is closer to 7-9 C. This allows for better detecfion of mismatches. Similarly, due to their non-ionic nature, hybridization of the bases attached to these backbones is relatively insensitive to salt concentration.
The nucleic acids may be single stranded or double stranded, as specified, or contain porbons of both double stranded or single stranded sequence. The nucleic acid may be DNA, both genomic and cDNA, RNA or a hybrid, where the nucleic acid contains any combination of deoxyribo- and ribo-nucleotides, and any combination of bases, including uracil, adenine, thymine, cytosine, guanine, inosine, xathanine hypoxathanine, isocytosine, isoguanine, etc. A preferred embodiment ufilizes isocytosine and isoguanine in nucleic acids designed to be complementary to other probes, rather than target sequences, as this reduces non-specific hybridizafion, as is generally described in U.S. Patent No. 5,681,702. As used herein, the term "nucleoside" includes nucleotides as well as nucleoside and nucleotide analogs, and modified nucleosides such as amino modified nucleosides. In addition, "nucleoside" includes non-naturally occuring analog structures. Thus for example the individual units of a peptide nucleic acid, each containing a base, are referred to herein as a nucleoside.

The composi6ons and methods of the invention are directed to the detecfion of target sequences. The term "target sequence" or "target nucleic acid" or grammatical equivalents herein means a nucleic acid sequence on a single strand of nucleic acid. The target sequence may be a por8on of a gene, a regulatory sequence, genomic DNA, cDNA, RNA including mRNA and rRNA, or others. As is outlined herein, the target sequence may be a target sequence from a sample, or a secondary target such as a product of a reaction such as a detection sequence from an invasive cleavage reaction, a ligated probe from an OLA reacfion, an extended probe from a PCR or SBE reaction, etc.
Thus, for example, a target sequence from a sample is amplified to produce a secondary target that is detected;
alternatively, an amplification step is done using a signal probe that is amplified, again producing a secondary target that is detected. The target sequence may be any length, with the understanding that longer sequences are more specific. As will be appreciated by those in the art, the complementary target sequence may take many forms. For example, it may be contained within a larger nucleic acid sequence, i.e. all or part of a gene or mRNA, a restriction fragment of a plasmid or genomic DNA, among others. As is outlined more fully below, probes are made to hybridize to target sequences to determine the presence, absence or quantity of a target sequence in a sample.
Generally speaking, this term will be understood by those skilled in the art.
The target sequence may also be comprised of different target domains; for example, in "sandwich" type assays as outlined below, a first target domain of the sample target sequence may hybridize to a capture probe or a portion of capture extender probe, a second target domain may hybridize to a por6on of an amplifier probe, a label probe, or a different capture or capture extender probe, etc.
In addition, the target domains may be adjacent (i.e. contiguous) or separated. For example, when OLA
techniques are used, a first primer may hybridize to a first target domain and a second primer may hybridize to a second target domain; either the domains are adjacent, or they may be separated by one or more nucleofides, coupled with the use of a polymerase and dNTPs, as is more fully outlined below. The terms "first" and "second" are not meant to confer an orientafion of the sequences with respect to the 5'-3' orientafion of the target sequence. For example, assuming a 5'-3' orientafion of the complementary target sequence, the first target domain may be located either 5' to the second domain, or 3' to the second domain. In addition, as will be appreciated by those in the art, the probes on the surface of the array (e.g. attached to the microspheres) may be attached in either orientation, either such that they have a free 3' end or a free 5' end; in some embodiments, the probes can be attached at one ore more internal posifions, or at both ends.

If required, the target sequence is prepared using known techniques. For example, the sample may be treated to lyse the cells, using known lysis buffers, sonication, electroporafion, etc., with purification and amplification as outlined below occurring as needed, as will be appreciated by those in the art. In addition, the reactions outlined herein may be accomplished in a variety of ways, as will be appreciated by those in the art. Components of the reaction may be added simultaneously, or sequentially, in any order, with preferred embodiments outlined below. In addifion, the reaction may include a variety of other reagents which may be included in the assays. These include reagents like salts, buffers, neutral proteins, e.g. albumin, detergents, etc., which may be used to facilitate optimal hybridization and detection, and/or reduce non-specific or background interactions. Also reagents that otherwise improve the efficiency of the assay, such as protease inhibitors, nuclease inhibitors, anti-microbial agents, etc., may be used, depending on the sample preparation methods and purity of the target.

In addition, in most embodiments, double stranded target nucleic acids are denatured to render them single stranded so as to permit hybridization of the primers and other probes of the invenfion. A
preferred embodiment utilizes a thermal step, generally by raising the temperature of the reaction to about 95 C, although pH changes and other techniques may also be used.

As outlined herein, the invention provides a number of different primers and probes. By "primer nucleic acid" herein is meant a probe nucleic acid that will hybridize to some pordon, i.e. a domain, of the target sequence. Probes of the present invenfion are designed to be complementary to a target sequence (either the target sequence of the sample or to other probe sequences, as is described below), such that hybridizafion of the target sequence and the probes of the present invention occurs.
As outlined below, this complementarity need not be perfect; there may be any number of base pair mismatches which will interfere with hybridizafion between the target sequence and the single 50.913-4 stranded nucieic acids of the present invention. However, if the number of mutations is so great that no hybridization can occur under even the least stringent of hybridizatlon conditlons, the sequence is not a complementary target sequence. Thus, by'substantially complementary' herein is meant that the probes are sufficiently complementary to the target sequences to hybridize under normal reacbon conditions.

A variety of hybridizatjon condittons may be used in the present invention, including high, moderate and low stringency condi6ons; see for example Maniatis et al., Molecular Cloning: A Laboratory Manual, 2d Edition,1989, and Short Protocols In Molecular Biology, ed.
Ausubel, at al.
Stringent conditions are sequence-dependent and will be different in different circumstances. Longer sequences hybridize speafically at higher temperetures. An extensive guide to the hybridization of nucleic adds Is found in Tijssen, Techniques In Biochemistry and Molecular Biology-Hybtidization with Nucleic Acid Probes, 'Overview of principies of hybridization and the strategy of nucleic add assays' (1993). Generally, stringent condidons are selected to be about 5-10'C lower than the thermal melting point (rm) for the specific sequence at a defined Ionic strength and pH. The Tm is the temperature (under defined ionic strength, pH
and nucleic acid concentration) at which 50% of the probes complementary to the target hybridize to the target sequence at equilibrium (as the target sequences are present in excess, at Tm, 50% of the probes are occupied at equiiibrium). Stringent conditions will be those in which the salt concentration is less than about 1.0 M sodium ion, typically about 0.01 to 1.0 M sodium ion concentration (or other satts) at pH
7.0 to 8.3 and the temperature is at least about 30'C for short probes (e.g.10 to 50 nucleotides) and at least about 60'C for long probes (e.g. greater than 50 nucleotides).
Stringent conditions may also be achieved with the additlon of hel'a destabilizing agents such as formamide.
The hybridization conditions may also vary when a non-ionic backbone, i.e. PNA is used, as Is known In the art In addiGon, cross-linking agents may be added after target binding to cross-link, i.e. covalently attach, the two strands of the hybridization complex.

Thus, the assays are generally run under stringency conditions which allows formation of the hybridization complex only in the presence of target. Stringency can be controlled by aitering a step parameter that is a thermodynamic variabie, inciuding, but not limited to, temperature, formamide concentration, salt concentration, chootropic salt concentration, pH, organic solvent concentration, etc.

These parameters may also be used to control non-specific binding, as is generally outlined in U.S.
Patent No. 5,681,697. Thus it may be desirable to perform certain steps at higher stringency conditions to reduce non-speclfic binding.

The size of the primer nucleic acid may vary, as will be appreciated by those in the art, in general varying from 5 to 500 nucleotides in length, with primers of between 10 and 100 being preferred, between 15 and 50 being particularly preferred, and from 10 to 35 being especially preferred, 509,13-4 depending on the use and amplification technique.

In addiiion, the different amplification techniques may have further requirements of the primers, as is more fully described below.

In addition, as outlined herein, a variety of labeling techniques can be done.
Labelina techniaues In general, either direct or indirect detection of the target products can be done. 'DirecY' detection as used In this context, as for the other reactlons outlined herein, requires the incorporation of a label, in this case a detectable label, preferably an op6cal label such as a fluorophore, Into the target sequence, with detection proceeding as outlined below. In this embodiment, the label(s) may be Incorporated in a variety of ways: (1) the primers comprise the label(s), for example attached to the base, a dbose, a phosphate, or to analogous structures in a nucleic acid analog; (2) modified nucleosides are used that are modified at either the base or the ribose (or to analogous structures In a nucleic acid analog) with the label(s); these label-modified nucleosides are then converted to the triphosphate form and are incorporated into a newly synthesized strand by a polymerase; (3) modified nucleotides are used that comprise a functional group that can be used to add a detectable label; (4) modified primers are used that comprise a func6onal group that can be used to add a detectable label or (5) a label probe that is directly labeled and hybridizes to a portion of the target sequence can be used. Any of these methods result in a newly synthesized strand or reaction product that comprises labels, that can be directly detected as outlined below.

Thus, the modified strands comprise a detec6on label. By'detection label' or'detectable label" herein is meant a moiety that allows detec6on. This may be a primary label or a secondary label.
Accordingly, detection labels may be primary labels (i.e. directiy detectable) or secondary labels (indirectly detectable).

In a preferred embodiment, the detection label is a primary label. A primary label is one that can be directly detected, such as a fluorophore. In general, labels fall into three classes: a) Isotopic labels, which may be radioactive or heavy isotopes; b) magnetic, electrical, thermal labels; and c) colored or luminescent dyes. Labels can also include enzymes (horseradish peroxidase, etc.) and magnetic particles. Preferred labels include chromophores or phosphors but are preferably fluorescent dyes.
Suitable dyes for use in the invention include, but are not limited to, fluorescent lanthanide complexes, including those of Europium and Terbium, fluorescein, rhodamine, tetramethylrhodamine, eosin, erythrosin, coumarin, methyl-coumarins, quantum dots (also referred to as'nanocrystals" ), pyrene, Malacite green, stilbene, Lucifer Yellow, Cascade BIueT"', Texas Red, Cy dyes (Cy3, Cy5, etc.), alexa dyes, phycoerytfiin, bodipy, and others described in the 6th Edition of the Molecular Probes Handbook by Richard P. Haugland, In a preferred embodiment, a secondary detectable label is used. A secondary label is one that is indirectly detected; for example, a secondary label-can bind or react with a primary label for detection, can act on an additional product to generate a primary label (e.g. enzymes), or may allow the separation of the compound comprising the secondary label from unlabeled materials, etc. Secondary labels find pardcular use in systems requiring separation of labeled and unlabeled probes, such as SBE, OLA, invasive cleavage reactions, etc; in addition, these techniques may be used with many of the other techniques descnbed herein. Secondary labels include, but are not limited to, one of a binding partner pair, chemically modifiable moieties; nuclease inhibitors, enzymes such as horseradish peroxidase, alkaline phosphatases, lucifierases, etc.

In a preferred embodiment, the secondary label is a binding partner pair. For example, the label may be a hapten or antigen, which will bind its binding partner. In a preferred embodiment, the binding partner can be attached to a solid support to allow separatlon of extended and non-extended primers.
For example, suitabie binding partner pairs Include, but are not limited to:
antigens (such as proteins (nciuding pep6ides)) and antibodies (including fragments thereof (FAbs, etc.)); proteins and small molecules, including biotin/streptavidin; enzymes and substrates or inhibitors; other protein-protein interac6ng pairs; receptor-ligands; and carbohydrates and their binding partners. Nucleic acid -nucieic acid binding proteins pairs are also useful. In general, the smaller of the pair is attached to the NTP for incorporation into the primer. Preferred binding partner pairs include, but are not limited to, biotin (or imino-biotin) and streptavidin, digeoidnin and Abs, and ProlinxTM' reagents (see www.prolinxinc.corr4e4/home.hmtl).

In a preferred embodiment, the binding partner pair comprises biofin or imino-biotln and streptavklin.
Imino-biotin is particularly preferred as imino-biotin disassociates from streptavidin In pH 4.0 buffer while biotin requires harsh denaturants (e.g. 6 M guanidinium HCI, pH 1.5 or 90% formamide at 95 C).

In a preferred embodiment, the binding partner pair comprises a primary detection label (for example, attached to the NTP and therefore to the extended primer) and an antibody that will specifically bind to the primary detection label. By `specifically bind" herein is meant that the partners bind with specificiiy sufficient to differentiate between the pair and other components or contaminants of the system. The binding should be sufficient to remain bound under the conditions of the assay, inciuding wash steps to remove non-specific binding. in some embodiments, the d'issociation constants of the pair will be less than about 10-4-10-0 M4, with less than about 10'5 to 10-0 M'' being preferred and less than about 10'' -10' M'' being particularly preferred.

In a preferred embodiment, the secondary label is a chemically modifiable moiety. In this embodiment, labels comprising reactive func6onal groups are Incorporated into the nucleic acid. The 50,913-4 funcdonal group can then be subsequently labeled with a primary label.
Suitable functional groups inciude, but are not limited to, amino groups, carboxy groups, maleimide groups, oxo groups and thiol groups, with amino groups and thiol groups being particularly preferred. For example, primary labels containing amino groups can be attached to secondary labels comprising amino groups, for example .5 using linkers as are known in the art; for example, homo-or hetero-bifunctional linkers as are well known (see 1994 Pierce Chemical Company catalog, technical secction on cross-linkers, pages 155-200).

For removal of unextended primers, it is preferred that the other half of the binding pair is attached to a solid support. In this embodiment, the solid support may be any as described herein for substrates and microspheres, and the form is preferably microspheres as well; for example, a preferred embodiment utilizes magnetic beads that can be easily introduced to the sample and easily removed, although any affinity chromatography formats may be used as well. Standard methods are used to attach the binding partner to the solid support, and can include direct or indirect attachment methods. For example, biotin labeled antibodies to fluorophores can be attached to streptavidin coated magnetk:
beads.

Thus, in this embodiment, the extended primers comprise a binding partner that is contacted with its binding partner under conditions wherein the extended or reacted primers are separated from the unextended or unreacted primers. These modified primers can then be added to the array comprising capture probes as described herein.

Removal of unextended primers In a preferred embodiment, it is desirable to remove the unextended or unreacted primers from the assay mixture, and particularly from the array, as unextended primers wiil compete with the extended (labeled) primers in binding to capture probes, thereby diminishing the signal. The concentration of the unextended primers relative to the extended primer may be relatively high, since a large excess of primer is usually required to generate efficient primer annealing.
Accordingly, a number of different techniques may be used to faciiitate the removal of unextended primers. While the discussion below applies specifically to SBE, these techniques may be used in any of the methods described herein.

In a preferred embodiment, the NTPs (or, in the case of other methods, one or more of the probes) comprise a secondary detectable label that can be used to separate extended and non-extended primers. As outlined above, detec6on labels may be primary labels (i.e.
directly detectable) or secondary labels (indirectly detectable). A secondary label is one that is indirectly detected; for example, a secondary label can bind or react with a primary label for detectlon, or may allow the separation of the compound comprising the secondary label from unlabeled materials, etc. Secondary labels find parflcular use in systems requiring separation of labeled and unlabeled probes, such as, SBE, OLA, invasive cleavage, etc. reactions; in addition, these techniques may be used with many of the other techniques described herein. Secondary labels include, but are not limited to, one of a binding partner pair; chemically modifiable moieties; nuclease inhibitors, etc.

In a preferred embodiment, the secondary label is a binding partner pair as outlined above. In a preferred embodiment, the binding partner pair comprises biotin or imino-biotin and streptavidin.
Imino-biotin is particularly preferred when the methods require the later separation of the pair, as imino-biotin disassociates from streptavidin in pH 4.0 buffer while biotin requires harsh denaturants (e.g. 6 M guanidinium HCI, pH 1.5 or 90% formamide at 95 C).

In addition, the use of streptavidin/biotin systems can be used to separate unreacted and reacted probes (for example in SBE, invasive cleavage, etc.). For example, the addition of streptavidin to a nucleic acid greatly increases its size, as well as changes its physical properties, to allow more efficient separation techniques. For example, the mixtures can be size fractionated by exclusion chromatography, afFinity chromatography, filtration or differential precipitation: Alternatively, an 3' exonuclease may be added to a mixture of 3' labeled biotin/streptavidin; only the unreacted oligonucleotides will be degraded. Following exonuclease treatment, the exonuclease and the streptavidin can be degraded using a protease such as proteinase K. The surviving nucleic acids (i.e.
those that were biotinylated) are then hybridized to the array.

In a preferred embodiment, the binding partner pair comprises a primary detection label (attached to the NTP and therefore to the extended primer) and an antibody that will specifically bind to the primary detection label.

In this embodiment, it is preferred that the other half of the binding pair is attached to a solid support.
In this embodiment, the solid support may be any as described herein for substrates and microspheres, and the form is preferably microspheres as well; for example, a preferred embodiment utilizes magnetic beads that can be easily introduced to the sample and easily removed, although any affinity chromatography formats may be used as well. Standard methods are used to attach the binding partner to the solid support, and can include direct or indirect attachment methods. For example, biotin labeled antibodies to fluorophores can be attached to streptavidin coated magnetic beads.

Thus, in this embodiment, the extended primers comprise a binding member that is contacted with its binding partner under conditions wherein the extended primers are separated from the unextended primers. These extended primers can then be added to the array comprising capture probes as described herein.

In a preferred embodiment, the secondary label is a chemically modifiable moiety. In this embodiment, labels comprising reactive funcfional groups are incorporated into the nucleic acid.

In a preferred embodiment, the secondary label is a nuclease inhibitor. In this embodiment, the chain-terminating NTPs are chosen to render extended primers resistant to nucleases, such as 3'-exonucleases. Addifion of an exonuclease will digest the non-extended primers leaving only the extended primers to bind to the capture probes on the array. This may also be done with OLA, wherein the ligated probe will be protected but the unprotected ligation probe will be digested.

In this embodiment, suitable 3'-exonucleases include, but are not limited to, exo I, exo III, exo VII, etc.
The present invenfion provides a variety of amplification reactions that can be detected using the arrays of the invention.

AMPLIFICATION REACTIONS

In this embodiment, the invention provides compositions and methods for the detection (and optionally quantification) of products of nucleic acid amplification reacfions, using bead arrays for detection of the amplification products. Suitable amplificafion methods include both target amplification and signal amplificafion and include, but are not limited to, polymerase chain reaction (PCR), ligation chain reaction (somefimes referred to as oligonucleotide ligase amplificafion OLA), cycling probe technology (CPT), strand displacement assay (SDA), transcripfion mediated amplification (TMA), nucleic acid sequence based amplification (NASBA), rolling circle amplification (RCA), and invasive cleavage technology. All of these methods require a primer nucleic acid (including nucleic acid analogs) that is hybridized to a target sequence to form a hybridization complex, and an enzyme is added that in some way modifies the primer to form a modified primer. For example, PCR generally requires two primers, dNTPs and a DNA polymerase; LCR requires two primers that adjacently hybridize to the target sequence and a ligase; CPT requires one cleavable primer and a cleaving enzyme; invasive cleavage requires two primers and a cleavage enzyme; etc. Thus, in general, a target nucleic acid is added to a reaction mixture that comprises the necessary amplification components, and a modified primer is formed.

In general, the modified primer comprises a detectable label, such as a fluorescent label, which is either incorporated by the enzyme or present on the original primer. As required, the unreacted primers are removed, in a variety of ways, as will be appreciated by those in the art and outlined herein. The hybridization complex is then disassociated, and the modified primer is detected and optionally quantitated by a microsphere array. In some cases, the newly modified primer serves as a target sequence for a secondary reaction, which then produces a number of amplified strands, which can be detected as outlined herein.

Accordingly, the reaction starts with the addifion of a primer nucleic acid to the target sequence which forms a hybridization complex. Once the hybridizatlon complex between the primer and the target.
sequence has been formed, an enzyme, sometimes termed an "ampiification enzyme", is used to modify the primer. As for all the methods outlined herein, the enzymes may be added at any point during the assay, either prior to, during, or after the addition of the primers. The identity of the enzyme will depend on the amplification technique used, as is more fully outlined below. Similarly, the modification will depend on the amplification technique, as outiined below.

Once the enzyme has modified the primer to form a modified primer, the hybridization complex is disassociated. In one aspect, dissociation is by modification of the assay conditions. In another aspect, the modified primer no longer hybridizes to the target nucleic acid and dissociates. Either one or both of these aspects can be employed in signal and target ampiification reac6ons as described below. Generally, the amplification steps are repeated for a period of time to allow a number of cycles, depending on the number of copies of the originai target sequence and the sensibvity of detection, with cycles ranging from 1 to thousands, with from 10 to 100 cycles being preferred and from 20 to 50 cycles being especially preferred.

After a suitable time of amplification, unreacted primers are removed, in a variety of ways, as will be appreciated by those in the art and described below, and the hybridization complex Is disassociated.
In general, the modified primer comprises a detectable label, such as a fluorescent label, which is either incorporated by the enzyme or present on the original primer, and the modified primer is added to a microsphere array such is generally described in PCT
applications US 98/09163 and US 99/14387; US 98/21193;
US 99/04473 and US 98/05025. The microsphere array comprises subpopulations of microspheres that comprise capture probes that wiii hybridize to the modified primers. Detection proceeds via detection of the label as an indication of the presence, absence or amount of the target sequence, as Is more fully oudined below.
TARGET AMPLIFICATION
In a preferred embodiment, the ampiification is target ampiification. Target amplification involves the amplification (repiication) of the target sequence to be detected, such that the number of copies of the target sequence is increased. Suitable target ampiification techniques Include, but are not limited to, the poiymerase chain reaction (PCR), strand displacement ampiification (SDA), transcription mediated amplification (TMA) and nucleic acid sequence based ampiification (NASBA).
POLYMERASE CHAIN REACTION AMPLIFICATION
In a preferred embodiment, the target amplification technique is PCR. The polymerase chain reaction (PCR) is wideiy used and described, and involves the use of primer extension combined with thermal cycling to amplify a target sequence; see U.S. Patent Nos. 4,683,195 and 4,683,202, and PCR
Essential Data, J. W. Wiley & sons, Ed. C.R. Newton, 1995.

In addition, there are a number of variations of PCR which also find use In the invention, inciuding 'quantitative competitive PCR' or'QC-PCR", 'arbitrariiy primed PCR" or "AP-PCR',"immuno-PCR', "Alu-PCR", "PCR single strand conformational polymorphism' or "PCR-SSCP', "reverse transcriptase PCR" or "RT-PCR", "biotin capture PCR", 'vectorette PCR', "panhandie PCR', and 'PCR select cDNA
subtraction", "alieie-spedfic PCR', among others. In some embodiments, PCR is not preferred.
In general, PCR may be briefly described as follows. A double stranded target nucieic acid is denatured, generally by raising the temperature, and then cooled in the presence of an excess of a PCR primer, which then hybridizes to the first target strand. A DNA polymerase then acts to extend the primer with dNTPs, resuiting In the synthesis of a new strand forming a hybridization complex.
The sample is then heated again, to disassociate the hybridizafion compiex, and the process is repeated. By using a second PCR primer for the complementary target strand, rapid and exponentiai ampiification occurs. Thus PCR steps are denaturation, annealing and extension. The partlcuiars of PCR are well known, and include the use of a thermostable polymerase such as Taq I polymerase and thermal cycling.

Accordingly, the PCR reaction requires at least one PCR primer, a polymerase, and a set of dNTPs.
As outlined herein, the primers may comprise the label, or one or more of the dNTPs may comprise a label.

In general, as is more fully outlined below, the capture probes on the beads of the array are designed to be substantiaily complementary to the extended part of the primer; that is, unextended primers wiil not bind to the capture probes. Altematively, as further described below, unreacted probes may be removed prior to addition to the array.

STRAND DISPLACEMENT AMPLIFICATION (SDA) In a preferred embodiment, the target ampiification technique Is SDA. Strand displacement amplification (SDA) is generally described in Walker et al., in Molecular Methods for Virus Detection, Academic Press, Inc., 1995, and U.S. Patent Nos. 5,455,166 and 5,130,238.

In general, SDA may be described as follows. A single stranded target nucieic add, usually a DNA
target sequence, is contacted with an SDA primer. An 'SDA primer' generally has a length of 25-100 nucieotides, with SDA primers of approximately 35 nucieotides being preferred.
An SDA primer is substantiaity complementary to a region at the 3' end of the target sequence, and the primer has a sequence at its 5' end (outside of the region that Is complementary to the target) that is a recognition sequence for a restriction endonuclease, somefimes referred to herein as a 'nicking enzytne' or a "nicking endonuclease", as outlined below. The SDA primer then hybridizes to the target sequence.
The SDA reaction mixture also contains a polymerase (an 'SDA polymerase', as outlined below) and a mbcture of all four deoxynucieoside-triphosphates (also called deoxynucleotides or dNTPs, i.e. dATP, dTTP, dCTP and dGTP), at least one species of which Is a substituted or modifled dNTP; thus, the SDA primer is modified, i.e. extended, to form a modified primer; sometimes referred to herein as a "newiy synthesized strand". The substituted dNTP is modified such that it will inhibit cleavage In the strand containing the substituted dNTP but will not inhibit cleavage on the other strand. Examples of suitabie substituted dNTPs include, but are not iimited, 2'deoxyadenosine 5'-O-(1-thiotriphosphate), 5-methyideoxycyddine 5'-triphosphate, 2'-deoxyuridine 5'-triphosphate, adn 7-deaza 2'-deoxyguanosine 5=triphosphate. In addition, the substitution of the dNTP may occur after incorporation into a newiy synthesized strand; for example, a methylase may be used to add methyl groups to the synthesized strand. In addi6on, if all the nucieotldes are substituted, the polymerase may have 5'-3' exonuciease activity. However, if less than all the nucleofides are substituted, the polymerase preferably lacks 5'y3' exonuclease activity.

As will be appreciated by those in the art, the recognition site/endonuclease pair can be any of a wide variety of known combinations. The endonuclease is chosen to cleave a strand either at the recognition site, or either 3' or 5' to it, without cleaving the compiementary sequence, either because the enzyme only cleaves one strand or because of the incorporation of the substituted nucleotides.
Suitable recognition site%ndonuclease pairs are well known in the art;
suitable endonucleases include, but are not fimited to, Hincll, Hindll, Aval, Fnu4HI, Tthilll, Ncll, BstXl, BamHl, etc. A chart depictlng suitable enzymes, and their corresponding recognition sites and the modified dNTP to use is found in U.S. Patent No. 5,455,166.

Once nicked, a polymerase (an "SDA polymerase") is used to extend the newly nicked strand, 5'=3', thereby creating another newly synthesized strand. The polymerase chosen should be able to intiate 5'--3' pofymerization at a nick site, should also displace the polymerized strand downstream from the nick, and should lack 5'-3' exonuciease actnriiy (this may be additionally accomplished by the additlon of a blocking agent). Thus, suitable polymerases in SDA include, but are not fimited to, the iqenow fragment of DNA polymerase I, SEQUENASE 1.0 and SEQUENASE 2.0 (U.S.
Biochemical), T5 DNA
polymerase and Phi29 DNA polymerase.

Accordingly, the SDA reac6on requires, in no particular order, an SDA primer, an SDA polymerase, a nicking endonuclease, and dNTPs, at least onespecies of which is modified.
Again, as outlined above for PCR, preferred embodiments utilize capture probes complementary to the newiy synthesized porBon of the primer, rather than the primer region, to allow unextended primers to be removed.

In general, SDA does not require thermocyciing. The temperature of the reaction is generally set to be high enough to prevent non-specific hybridization but low enough to allow spedfic hybridization; this I,s generally from about 37'C to about 42'C, depending on the enzymes.

In a preferred embodiment, as for most of the amplification techniques described herein, a second amplification reaction can be done using the complementary target sequence, resulUng in a substantial increase in ampfification during a set period of time. That is, a second primer nucleic acid Is hybridized to a second target sequence, that is substantially complementary to the first target sequence, to form a second hybridization complex. The addition of the enzyme, followed by disassociation of the second hybridization complex, results in the generation of a number of newly synthesized second strands.

NUCLEIC ACID SEQUENCE BASED AMPLIFICATION (NASBA) AND TRANSCRIPTION MEDIATED
AMPLIFICATION (TMA) In a preferred embodiment, the target amplifica6on technique is nucleic acid sequence based amplification (NASBA). NASBA is generally described in U.S. Patent No.
5,409,818; Sooknanan et al., Nucleic Acid Sequence-Based Ampiificatlon, Ch. 12 (pp. 261-285) of Molecular Methods for Virus Detection, Academic Press, 1995; and "Profiting from Gene-based Diagnostirs', CTB Intemational Publishing Inc., N.J., 1996, all of which are incorporated by reference. NASBA
is very similar to both TMA and QBR. Transcription mediated amplification (TMA) is generally described In U.S. Patent Nos. 5,399,491, 5,888,779, 5,705,365, 5,710,029. The main difference between NASBA and TMA is that NASBA utilizes the addition of RNAse H to effect RNA
degradation, and TMA relies on inherent RNAse H aclivity of the reverse transcriptase.

In general, these techniques may be desctibed as follows. A single stranded target nucleic acid, usually an RNA target sequence (sometimes referred to herein as 'the first target sequence' or "the first templatel, is contacted with a first primer, generally referred to herein as a"NASBA primer"
(although "TMA primer" Is also suitable). Starfing with a DNA target sequence is described below.
These primers generally have a length of 25-100 nucleotides, with NASBA
primers of approximately 50-75 nucleotides being preferred. The first primer is preferably a DNA primer that has at its 3' end a sequence that is substantlally complementary to the 3' end of the first template. The first primer also has an RNA polymerase promoter at its 5' end (or its complement (antisense), depending on the configuration of the system). The first primer Is then hybridized to the first template to form a first hybridization complex. The reactlon mature also includes a reverse transcriptase enzyme (an "NASBA reverse transcriptase") and a mbcture of the four dNTPs, such that the first NASBA primer is modified, i.e. extended, to form a modified first primer, comprising a hybridization complex of RNA (the first template) and DNA (the newly synthesized strand).

By "reverse transcriptase" or "RNA-directed DNA polymerase" herein is meant an enzyme capable of synthesizing DNA from a DNA primer and an RNA template. Su'dable RNA-directed DNA
polymerases Include, but are not limited to, avian myloblastosis virus reverse transcriptase ("AMV RT") and the Moloney murine leukemia virus RT. When the amplification reaction is TMA, the reverse transcriptase enzyme further comprises a RNA degrading activity as outlined below.

In addition to the components listed above, the NASBA reaction also includes an RNA degrading enzyme, also sometimes referred to herein as a ribonuclease, that will hydrolyze RNA of an RNA:DNA
hybrid without hydrolyzing single- or double-stranded RNA or DNA. Suitable ribonucleases include, but are not limited to, RNase H from E. coli and calf thymus.

The ribonuclease activity degrades the first RNA template in the hybridization complex, resulting in a disassociation of the hybridizafion complex leaving a first single stranded newly synthesized DNA
strand, sometimes referred to herein as "the second template".

In addition, the NASBA reaction also includes a second NASBA primer, generally comprising DNA
(although as for all the probes herein, including primers, nucleic acid analogs may also be used). This second NASBA primer has a sequence at its 3' end that is substanfially complementary to the 3' end of the second template, and also contains an antisense sequence for a functional promoter and the antisense sequence of a transcription inifiation site. Thus, this primer sequence, when used as a template for synthesis of the third DNA template, contains sufficient information to allow specific and efficient binding of an RNA polymerase and initiafion of transcription at the desired site. Preferred embodiments utilizes the antisense promoter and transcription inifiation site are that of the T7 RNA
polymerase, although other RNA polymerase promoters and initiation sites can be used as well, as outlined below.

The second primer hybridizes to the second template, and a DNA polymerase, also termed a "DNA-directed DNA polymerase", also present in the reaction, synthesizes a third template (a second newly synthesized DNA strand), resulting in second hybridization complex comprising two newly synthesized DNA strands.

Finally, the inclusion of an RNA polymerase and the required four ribonucleoside triphosphates (ribonucleotides or NTPs) results in the synthesis of an RNA strand (a third newly synthesized strand that is essentially the same as the first template). The RNA polymerase, somefimes referred to herein as a "DNA-directed RNA polymerase", recognizes the promoter and specifically inifiates RNA
synthesis at the initiation site. In addition, the RNA polymerase preferably synthesizes several copies of RNA per DNA duplex. Preferred RNA polymerases include, but are not limited to, T7 RNA
polymerase, and other bacteriophage RNA polymerases including those of phage T3, phage (~II, Salmonella phage sp6, or Pseudomonase phage gh-1.

In some embodiments, TMA and NASBA are used with starbng DNA target sequences.
In this embodiment, it is necessary to utilize the first primer comprising the RNA
polymerase promoter and a DNA polymerase enzyme to generate a double stranded DNA hybrid with the newly synthesized strand comprising the promoter sequence. The hybrid is then denatured and the second primer added.

Accordingly, the NASBA reaction requires, in no par6cular order, a first NASBA
primer, a second NASBA primer comprising an antisense sequence of an RNA polymerase promoter, an RNA
polymerase that recognizes the promoter, a reverse transctiptase, a DNA
polymerase, an RNA
degrading enzyme, NTPs and dNTPs, in addition to the detection components outlined below.

These components result in a single star6ng RNA template generafing a single DNA duplex; however, since this DNA duplex results in the creafion of muitiple RNA strands, which can then be used to initiate the reac6on again, amplification proceeds rapidly.

Accordingly, the TMA reacfion requires, in no particular order, a first TMA
primer, a second TMA
primer comprising an antisense sequence of an RNA polymerase promoter, an RNA
polymerase that recognizes the promoter, a reverse transcriptase with RNA degrading activity, a DNA polymerase, NTPs and dNTPs, in addifion to the detection components outlined below.

These components result in a single starting RNA template generating a single DNA duplex; however, since this DNA duplex results in the creation of multiple RNA strands, which can then be used to initiate the reaction again, amplification proceeds rapidly.

As outlined herein, the detection of the newly synthesized strands can proceed in several ways. Direct detecfion can be done when the newly synthesized strands comprise detectable labels, either by incorporation into the primers or by incorporation of modified labelled nucleotides into the growing strand. Alternatively, as is more fully outlined below, indirect detection of unlabelled strands (which now serve as "targets" in the detection mode) can occur using a variety of sandwich assay configurations. As will be appreciated by those in the art, any of the newly synthesized strands can serve as the "target" for form an assay complex on a surface with a capture probe. In NASBA and TMA, it is preferable to utilize the newly formed RNA strands as the target, as this is where significant amplificafion occurs.

In this way, a number of secondary target molecules are made. As is more fully outlined below, these reactions (that is, the products of these reacfions) can be detected in a number of ways.

SIGNAL AMPLIFICATION TECHNIQUES
In a preferred embodiment, the amplification technique is signal amplificafion. Signal amplification involves the use of limited number of target molecules as templates to either generate multiple signalling probes or allow the use of multiple signalling probes. Signal amplification strategies include LCR, CPT, Q(3R, invasive cleavage technology, and the use of amplification probes in sandwich assays.

SINGLE BASE EXTENSION (SBE) In a preferred embodiment, single base extension (SBE; sometimes referred to as'minisequencing") is used for ampiification. It should also be noted that SBE finds use In genotyping, as is described below. Briefly, SBE is a technique that utilizes an extension primer that hybridizes to the target nucleic acid. A polymerase (generally a DNA poiymerase) is used to extend the 3' end of the primer with a nucieotide analog labeled a detection label as described herein. Based on the fidelity of the enzyme, a nucieotide is only incorporated into the extension primer if it Is complementary to the adjacent base in the target strand. Generally, the nucleotide Is derivatized such that no further extensions can occur, so only a single nucleotide is'added. However, for ampiification reactions, this may not be necessary.
Once the labeled nucleotide is added, detection of the label proceeds as outlined herein. See generally Syivanen et al., Genomics 8:684-692 (1990); U.S. Patent Nos.
5,846,710 and 5,888,819;
Pas6nen et al., Genomics Res. 7(6):606-614 (1997).

The reaction is initiated by introducing the assay complex comprising the target sequence (i.e. the array) to a solution comprising a first nucleotide, frequently an nucieotide analog. By "nucieotide analog" in this context herein is meant a deoxynucleoside-triphosphate (also called deoxynucieotides or dNTPs, i.e. dATP, dTTP, dCTP and dGTP), that Is further derivatized to be chain terminating. As will be appreciated by those in the art, any number of nucleotide analogs may be used, as long as a polymerase enzyme will stili incorporate the nucleotide at the interrogation position. Preferred embodiments utilize dideoxy-triphosphate nucleotides (ddNTPs). Generally, a set of nucieotides comprising ddATP, ddCTP,, ddGTP and ddTTP is used, at least one of which Includes a label, and preferably all four. F.or ampification rather than genotyping reac6ons, the labels may all be the same;
altemativeiy, different labels may be used.

In a preferred embodiment, the nucleotide analogs comprise a detectable label, which can be either a primary or secondary detectable label. Preferred primary labels are those outiined above. However, the enzymatic incorporation of nucleotides comprising fluorophores is poor under many condi6ons;
accordingly, preferred embodiments utilize secondary detectable labels. In addition, as outlined below, the use of secondary labels may also facilitate the removal of unextended probes.

In addition to a first nucieotide, the soiution also comprises an extension enzyme, generally a DNA
polymerase. Suitable DNA polymerases include, but are not limited to, the Kienow fragment of DNA
poiymerase I, SEQUENASE 1.0 and SEQUENASE 2.0 (U.S. Biochemical), T5 DNA
polymerase and Phi29 DNA polymerase. If the NTP is complementary to the base of the detection position of the target sequence, which is adjacent to the extension primer, the extension enzyme will add itto the extension primer. Thus, the extension primer is modified, i.e. extended, to form a modified primer, sometimes referred to herein as a"newiy synthesized strand".

A limitation of this method is that unless the target nucleic acid is In sufficient concentration, the 50,913-4 amount of unextended primer in the reaction greatiy exceeds the resultant extended-labeled primer.
The excess of unextended primer competes with the detection of the labeled primer In the assays described herein. Accordingly, when SBE is used, preferred embodiments utiiize methods for the removal of unextended primers as outlined herein.

One method to overcome this iimitation is thermocyciing minisequencing in which repeated cycles of annealing, primer extension, and heat denaturation using a thermocycler and thermo-shable polymerase allows the ampiification of the extension probe which resuits in the accumuiation of extended primers. For example, if the original unextended primer to target nucleic acid concentration is 100:1 and 100 thermocycles and extensiona are performed, a majority of the primer will be extended.

As wiii be appreciated by those in the art, the configuration of the SBE
system can take on several forms. As for the LCR reac6on described below, the reaction may be done in soiution, and then the newly synthesized strands, with the base-specific detectable labels, can be detected. For example, they can be directiy hybridized to capture probes that are complementary to the extension primers, and the presence of the label is then detected.

Alternatively, the SBE reaction can occur on a surface. For example, a target nucieic acid may be captured using a first capture probe that hybridizes to a first target domain of the target, and the reactlon can proceed at a second target domain. The extended labeled primers are then bound to a second capture probe and detected.

Thus, the SBE reaction requires, in no particuiar order, an extension primer, a polymerase and dNTPs, at least one of which is labeled.

OLIGONUCLEOTIDE LIGATION AMPLIFICATION (OLA) In a preferred embodiment, the signal ampiification technique is OLA. OLA, which is referred to as the iigation chain reaction (LCR) when two-stranded substrates are used, involves the ligation of two smaller probes into a single long probe, using the target sequence as the template. In LCR, the ligated probe product becomes the predominant template as the reaction progresses. The method can be run in two different ways; in a first embodiment, only one strand of a target sequence is used as a template for iigation; afternativeiy, both strands may be used. See generally U.S. Patent Nos.
5,185,243, 5,679,524 and 5,573,907; EP 0 320 308 131; EP 0 336 731 B1; EP 0 439 182 61; WO
90/01069; WO 89/12696; WO 97/31256; and WO 89/09835.

In a preferred embodiment, the single-stranded target sequence comprises a first target domain and a second target domain, which are adjacent and contiguous. A first OLA primer and a second OLA

primer nucleic acids are added, that are substantially complementary to their respective target domain and thus will hybridize to the target domains. These target domains may be directly adjacent, i.e.
contiguous, or separated by a number of nucleotides. If they are non-contiguous, nucleotides are added along with means to join nucleotides, such as a polymerase, that will add the nucleotides to one of the primers. The two OLA primers are then covalently attached, for example using a ligase enzyme such as is known in the art, to form a modified primer. This forms a first hybridization complex comprising the ligated probe and the target sequence. This hybridization complex is then denatured (disassociated), and the process is repeated to generate a pool of ligated probes.

In a preferred embodiment, OLA is done for two strands of a double-stranded target sequence. The target sequence is denatured, and two sets of probes are added: one set as outlined above for one strand of the target, and a separate set (i.e. third and fourth primer probe nucleic acids) for the other strand of the target. In a preferred embodiment, the first and third probes will hybridize, and the second and fourth probes will hybridize, such that amplification can occur.
That is, when the first and second probes have been attached, the ligated probe can now be used as a template, in addition to the second target sequence, for the attachment of the third and fourth probes.
Similarly, the ligated third and fourth probes will serve as a template for the attachment of the first and second probes, in addition to the first target strand. In this way, an exponential, rather than just a linear, amplification can occur.

As will be appreciated by those in the art, the ligation product can be detected in a variety of ways. In a preferred embodiment, the ligation reaction is run in solution. In this embodiment, only one of the primers carries a detectable label, e.g. the first ligation probe, and the capture probe on the bead is substantially complementary to the other probe, e.g. the second ligation probe. In this way, unextended labeld ligation primers will not interfere with the assay. That is, in a preferred embodiment, the ligation product is detected by solid-phase oligonucleotide probes. The solid-phase probes are preferably complementary to at least a portion of the ligation product. In a preferred embodiment, the solid-phase probe is complementary to the 5' detection oligonucleotide portion of the ligation product.
This substantially reduces or eliminates false signal generated by the optically-labeled 3' primers.
Preferably, detection is accomplished by removing the unligated 5' detection oligonucleotide from the reaction before application to a capture probe. In one embodiment, the unligated 5' detection oligonucleotides are removed by digesting 3' non-protected oligonucleotides with a 3' exonuclease, such as, exonuclease I. The ligation products are protected from exo I
digestion by including, for example, 4-phosphorothioate residues at their 3' terminus, thereby, rendering them resistant to exonuclease digestion. The unligated detection oligonucleotides are not protected and are digested.
Alternatively, the target nucleic acid is immobilized on a solid-phase surface. The ligafion assay is performed and unligated oligonucleotides are removed by washing under appropriate stringency to remove unligated oligonucleotides. The ligated oligonucleotides are eluted from the target nucleic acid 50,913-4 using denaturing conditions, such as, 0.1 N NaOH, and detected as described herein.

Again, as outlined above, the detection of the LCR reaction can also occur directly; in the case where one or both of the primers comprises at least one detectable label, or indirectly, using sandwich assays, through the use of additional probes; that is, the ligated probes can serve as target sequences, and detection may utilize amplification probes, capture probes, capture extender probes, label probes, and label extender probes, etc.

ROLLING-CIRCLE AMPLIFICATION (RCA) In a preferred embodiment the signal amplification technique is RCA. Rolling-circle amplification is generally described in Baner et al. (1998) Nuc. Acids Res. 26:5073-5078;
Barany, F. (1991) Proc. Natl.
Acad. Sci. USA 88:189-193; and Uzardi et al. (1998) Nat. Genet.19:225-232, In general, RCA may be described In two ways. First, as is oudined in more detail below, a single probe is hybridized with a target nucleic acid. Each terminus of the probe hybridizes adjacently on the target nucleic acid and the OLA assay as described above occurs. When ligated, the probe Is circularized while hybridized to the target nucleic acid. Additlon of a polymerase resuits In extension of the circular probe. However, since the probe has no terminus, the polymerase continues to extend the probe repeatedly. Thus results in amplification of the circular probe.

A second alternative approach involves OLA followed by RCA. In this embodiment, an immobiiized primer Is contacted with a target nucleic acid. Compiementary sequences will hybridize with each other resulting in an immobilized duplex. A second primer is contacted with the target nucleic acid.
The second primer hybridizes to the target nucleic acid adjacent to the first primer. An OLA assay is performed as described above. Ugation only occurs if the primer are complementary to the target nucleic acid. When a mismatch occurs, particularly at one of the nucleotides to be ligated, ligation will not occur. Following figation of the oligonucleotides, the ligated, Immobilized, ol'igonucleotide Is then hybridized with an RCA probe. This is a circular probe that is designed to specificaliy hybridize with the ligated oligonucleotide and will only hybridize with an oligonucleotide that has undergone ligation.
RCA is then performed as Is outlined in more detail below.

Accordingly, in an preferred embodiment, a single oligonucleotide Is used both for OLA and as the circular template for RCA (referred to herein as a"padlock probe" or a'RCA
probe"). That is, each terminus of the oligonucleotide contains sequence complementary to the target nucleic acid and functions as an OLA primer as described above. That is, the first end of the RCA probe Is substantially complementary to a first target domain, and the second end of the RCA probe is substantially complementary to a second target domain, adjacent to the first domain. Hybridization of the oligonucleotide to the target nucleic acid results in the formation of a hybridization complex.

Ligation of the "primers" (which are the discrete ends of a single oligonucleofide) results in the formation of a modified hybridization complex containing a circular probe i.e.
an RCA template complex. That is, the oligonucleotide is circularized while still hybridized with the target nucleic acid.
This serves as a circular template for RCA. Addifion of a polymerase to the RCA template complex results in the formation of an amplified product nucleic acid. Following RCA, the amplified product nucleic acid is detected (Figure 6). This can be accomplished in a variety of ways; for example, the polymerase may incorporate labelled nucleofides, or alternatively, a label probe is used that is substantially complementary to a porbon of the RCA probe and comprises at least one label is used.
The polymerase can be any polymerase, but is preferably one lacking 3' exonuclease activity (3' exo ).
Examples of suitable polymerase include but are not limited to exonuclease minus DNA Polymerase I
large (Klenow) Fragment, Phi29 DNA polymerase, Taq DNA Polymerase and the like. In addition, in some embodiments, a polymerase that will replicate single-stranded DNA (i.e.
vAthout a primer forming a double stranded section) can be used.

In a preferred embodiment, the RCA probe contains an adapter sequence as outlined herein, with adapter capture probes on the array, for example on a microsphere when microsphere arrays are being used. Alternatively, unique por6ons of the RCA probes, for example all or part of the sequence corresponding to the target sequence, can be used to bind to a capture probe.

In a preferred embodiment, the padlock probe contains a restriction site. The restriction endonuclease site allows for cleavage of the long concatamers that are typically the result of RCA into smaller individual units that hybridize either more efficiently or faster to surface bound capture probes. Thus, following RCA, the product nucleic acid is contacted with the appropriate restriction endonuclease.
This results in cleavage of the product nucleic acid into smaller fragments.
The fragments are then hybridized with the capture probe that is immobilized resulting in a concentration of product fragments onto the microsphere. Again, as outlined herein, these fragments can be detected in one of two ways:
either labelled nucleotides are incorporated during the replication step, or an additional label probe is added.

Thus, in a preferred embodiment, the padlock probe comprises a label sequence;
i.e. a sequence that can be used to bind label probes and is substantially complementary to a label probe. In one embodiment, it is possible to use the same label sequence and label probe for all padlock probes on an array; alternatively, each padlock probe can have a different label sequence.

The padlock probe also contains a priming site for priming the RCA reaction.
That is, each padlock probe comprises a sequence to which a primer nucleic acid hybridizes forming a template for the polymerase. The primer can be found in any por6on of the circular probe. In a preferred embodiment, the primer is located at a discrete site in the probe. In this embodiment, the primer site in each distinct padlock probe is identical, although this is not required. Advantages of using primer sites with identical sequences include the ability to use only a single primer oligonucleotide to prime the RCA assay with a plurality of different hybridization complexes. That is, the padlock probe hybridizes uniquely to the target nucleic acid to which it is designed. A single primer hybridizes to all of the unique hybridization complexes forming a priming site for the polymerase. RCA then proceeds from an identical locus within each unique padlock probe of the hybridization complexes.

In an alternative embodiment, the primer site can overlap, encompass, or reside within any of the above-described elements of the padlock probe. That is, the primer can be found, for example, overlapping or within the restriction site or the identifier sequence. In this embodiment, it is necessary that the primer nucleic acid is designed to base pair with the chosen primer site.

Thus, the padlock probe of the invention contains at each terminus, sequences corresponding to OLA
primers. The intervening sequence of the padlock probe contain in no particular order, an adapter sequence and a restriction endonuclease site. In addition, the padlock probe contains a RCA priming site.

Thus, in a preferred embodiment the OLA/RCA is performed in solufion followed by restriction endonuclease cleavage of the RCA product. The cleaved product is then applied to an array comprising beads, each bead comprising a probe complementary to the adapter sequence located in the padlock probe. The amplified adapter sequence correlates with a particular target nucleic acid.
Thus the incorporation of an endonuclease site allows the generation of short, easily hybridizable sequences. Furthermore, the unique adapter sequence in each rolling circle padlock probe sequence allows diverse sets of nucleic acid sequences to be analyzed in parallel on an array, since each sequence is resolved on the basis of hybridizafion specificity.

In an alternative OLA/RCA method, one of the OLA primers is immobilized on the microsphere; the second primer is added in solution. Both primers hybridize with the target nucleic acid forming a hybridization complex as described above for the OLA assay.

As described herein, the microsphere is distributed on an array. In a preferred embodiment, a plurality of microspheres each with a unique OLA primer is distributed on the array.

Following the OLA assay, and either before, after or concurrently with distribution of the beads on the array, a segment of circular DNA is hybridized to the bead-based ligated oligonucleotide forming a modified hybridization complex. Addifion of an appropriate polymerase (3' exo-), as is known in the art, and corresponding reaction buffer to the array leads to amplification of the circular DNA. Since there is no terminus to the circular DNA, the polymerase continues to travel around the circular template generating extension product until it detaches from the template. Thus, a polymerase with high processivity can create several hundred or thousand copies of the circular template with all the copies linked in one configuous strand.

Again, these copies are subsequently detected by one of two methods; either hybridizing a labeled oligo complementary to the circular target or via the incorporation of labeled nucleotides in the amplification reacfion. The label is detected using convenfional label detecfion methods as described herein.

In one embodiment, when the circular DNA contains sequences complementary to the ligated oligonucleofide it is preferable to remove the target DNA prior to contacting the ligated oligonucleotide with the circular DNA (See Fig 7). This is done by denaturing the double-stranded DNA by methods known in the art. In an alternative embodiment, the double stranded DNA is not denatured prior to contacting the circular DNA.

In an alternative embodiment, when the circular DNA contains sequences complementary to the target nucleic acid, it is preferable that the circular DNA is complementary at a site distinct from the site bound to the ligated oligonucleotide. In this embodiment it is preferred that the duplex between the ligated oligonucleotide and target nucleic acid is not denatured or disrupted prior to the addition of the circular DNA so that the target DNA remains immobilized to the bead.

Hybridization and washing conditions are well known in the art; various degrees of stringency can be used. In some embodiments it is not necessary to use stringent hybridization or washing conditions as only microspheres containing the ligated probes will effectively hybridize with the circular DNA;
microspheres bound to DNA that did not undergo ligation (those without the appropriate target nucleic acid) will not hybridize as strongly with the circular DNA as those primers that were ligated. Thus, hybridization and/or washing condifions are used that discriminate between binding of the circular DNA
to the ligated primer and the unligated primer.

Alternatively, when the circular probe is designed to hybridize to the target nucleic acid at a site distinct from the site bound to the ligated oligonucleotide, hybridizafion and washing condifions are used to remove or dissociate the target nucleic acid from unligated oligonucleotides while target nucleic acid hybridizing with the ligated oligonucleotides will remain bound to the beads.
In this embodiment, the circular probe only hybridizes to the target nucleic acid when the target nucleic acid is hybridized with a ligated oligonucleotide that is immobilized on a bead.

As is well known in the art, an appropriate polymerase (3' exo ) is added to the array. The polymerase extends the sequence of a single-stranded DNA using double-stranded DNA as a primer site. In one embodiment, the circular DNA that has hybridized with the appropriate OLA
reaction product serves as the primer for the polymerase. In the presence of an appropriate reaction buffer as is known in the art, the polymerase will extend the sequence of the primer using the single-stranded circular DNA as a template. As there is no terminus of the circular DNA, the polymerase will continue to extend the sequence of the circular DNA. In an alternative embodiment, the RCA probe comprises a discrete primer site located within the circular probe. Hybridization of primer nucleic acids to this primer site forms the polymerase tempiate allowing RCA to proceed.

In a preferred embodiment, the polymerase creates more than 100 copies of the circular DNA. In more preferred embodiments the poiymerase creates more than 1000 copies of the circular DNA;
while in a most preferred embodiment the polymerase creates more than 10,000 copies or more than 50,000 copies of the template.

The amplified circular DNA sequence is then detected by methods known in the art and as described herein. Detection is accomplished by hybridizing with a labeled probe. The probe Is labeled directly or indirectly. Alternatively, labeled nucieotides are incorporated into the ampiified circular DNA product.
The nucieotides can be labeled directly, or indirectiy as is further described herein.

The RCA as described herein finds use in allowing highly specific and highly sensiifve detection of nucleic acid target sequences. In particuiar, the method finds use in improving the multiplexing ability of DNA arrays and eiiminating costly sample or target preparation. As an example, a substantlal savings in cost can be realized by directly analyzing genomic DNA on an array, rather than employing an intermediate PCR ampiification step. The method finds use in examining genomic DNA and other samples including mRNA.

In addition the RCA finds use in aiiowing rolling circle ampiification products to be easily detected by hybridization to probes in a solid-phase format (e.g. an array of beads). An additional advantage of the RCA is that it provides the capabiiity of muitipiex analysis so that large numbers of sequences can be analyzed in parallel. By combining the sensitivity of RCA and parallel detection on arrays, many sequences can be analyzed directly from genomic DNA.

CHEMICAL LIGATION TECHNIQUES
' A variation of LCR utilizes a"chemicai iigation' of sorts, as is generally outiined In U.S. Patent Nos. 5,616,464 and 5,767,259. In this embodiment, similar to enzymatic ligation, a pair of primers are utilized, wherein the first primer is substantially complementary to a first domain of the target and the second primer is substantiaiiy complementary to an adjacent second domain of the target (aithough, as for enzymatic iigation, if a"gap" exists, a polymerase and dNTPs may be added to'fdi in' the gap). Each primer has a portion that acts as a"side chain' that does not bind the target sequence and acts as one half of a stem structure that interacts non-covalently through hydrogen bonding, salt bridges, van der Waal's forces, etc. Preferred embodiments utiiize substantiaiiy complementary nucleic acids as the side 50,913-4 chains. Thus, upon hybridization of the primers to the target sequence, the side chains of the primers are brought into spatiai proximity, and, If the side chains comprise nucleic acids as well, can also form side chain hybridizafion complexes.

At least one of the side chains of the primers comprises an adivatabie cross-iinidng agent, generally covalently attached to the side chain, that upon activation, resutts In a chemical cross-tink or chemical iigation. The adivatibie group may comprise any moiety that will allow cross-linking of the side chains, and include groups activated chemically, photonically and thermally, with photoaCtivatabie groups being preferred. In some embodiments a single adivatabie group on one of the side chains is enough to result in cross-linidng via interaction to a functionai group on the other side chain; in altemate embodiments, activatabie groups are required on each side chain.

Once the hybridization complex is formed, and the cross-linking agent has been activated such that the primers have been covaientiy attached, the reaction is subjected to conditions to allow for the disassocation of the hybridization complex, thus freeing up the target to serve as a template for the next ligation or cross-linking. In this way, signal ampiification occurs, and can be detected as outlined herein.

INVASIVE CLEAVAGE TECHNIQUES
In a preferred embodiment, the signal amplfication technique is invasive cleavage technology, which Is described in a number of patents and patent appiications, induding U.S.
Patent Nos. 5,846,717;
5,614,402; 5,719,028; 5,541,311; and 5,843,669. Invasive deavage technology is based on structure-specific nucleases that cleave nucleic acids in a site-specific manner. Two probes are used: an "invader" probe and a"signalling"
probe, that adjacently hybridize to a target sequence with overlap. For mismatch discrimination, the invader technology relies on compiementarity at the overlap positlon where cleavage occurs. The enzyme cleaves at the overiap, and releases the "tail" which may or may not be labeled. This can then be detected.

Generally, invasive cleavage technology may be described as follows. A target nucieic acid is recognized by two distinct probes. A first probe, generally referred to herein as an "invader" probe, is substantially complementary to a first portion of the target nucleic acid. A
second probe, generally referred to herein as a "signal probe', is partially complementary to the target nucielc acid; the 3' end of the signal oligonucleotide is substantiaiiy complementary to the target sequence while the 5' end is non-complementary and preferably forms a single-stranded "tail" or "arm". The non-complementary end of the second probe preferably comprises a"generic" or "unique" sequence, frequently referred to herein as a"detection sequence", that is used to indicate the presence or absence of the target nucieic acid, as described below. The detection sequence of the second probe preferably comprises at least one detectable label, although as outiined herein, since this detecfion sequence can functlon as a 43.

target sequence for a capture probe, sandwich configurations utilizing label probes as described herein may also be done.

Hybridization of the first and second oligonudeotkles near or adjacent to one another on the target nucleic add forms a number of structures. In a preferred embodiment, a forked cleavage structure forms and Is a substrate of a nuclease which cleaves the detection sequence from the signal oligonucieotide. The site of cleavage Is controlled by the distance or overlap between the 3' end of the invader oiigonucleotkie and the downstream fork of the signal oligonucleotlde.
Therefore, neitlter oligonucleotide Is subject to cleavage when misaligned or when unattached to target nudeic add.

In a preferred embodiment, the nuciease that recognizes the forked cleavage structure and catalyzes release of the tail Is thermostable, thereby, aiiowing thermal cycling of the cleavage reaction, if desired. Preferred nucleases derived from thermostable DNA polymerases that have been modified to have reduced synthetic activity which is an undesirable side-reacbon during cleavage are disclosed in U.S. Patent Nos. 5,719,028 and 5,843,669. The synthetic activity of the DNA
polymerase is reduced to a level where it does not interfere with detecion of the deavage reaction and detection of the freed tall. Preferably the DNA poiymerase has no detectebie polymerase adivity. Examples of nucleases are those derived from Thermus aquadcus, Thennus flevus, or Thennus thermophllus.

In another embodiment, thermostable structure-specific nucleases are Flap endonucleases (FENs) selected from FEN-1 or FEN-2 like (e.g. XPG and RAD2 nucleases) from Archaebacteriai species, for example, FEN-1 from MeUianococcus jannascahl, Pyrococcus fiuiosfs, Pyrococcus woesei, and Archaeoglobus fulgidus. (U.S. Patent No. 5,843,669 and Lyamichev et aL 1999.
Nature Biotechnology 17:292-297).

In a preferred embodiment, the nuclease is AfuFENI or PfuFENI nuclease. To cleave a forked structure, these nucleases require at least one overlapping nudeotide between the signai and invasive probes to recognize and cleave the 5' end of the signal probe. To effect cleavage the 3'-terminal nucleotide of the Invader oligonucleotide Is not required to be complementary to the target nucieie add. In contast, mismatch of the signal probe one base upstream of the cleavage site prevents creation of the overiap and cleavage. The specificity of the nuclease reac6on allows single nucleotide poiymorphism (SNP) detection from, for example, genomic DNA, as outtined below (Lyamichev et al.).

The invasive cleavage assay Is preferably performed on an array format. In=a.
preferred embodiment, the signal probe has a detectable label, attached 5' from the site of nuclease cleavage (e.g. within the detection sequence) and a capture tag, as described below (e.g. biotin or other hapten) 3' from the site of nuclease cleavage. After the assay is carried out, the 3' portion of the cleaved signal probe (e.g. the the detecfion sequence) are extracted, for example, by binding to streptavidin beads or by crosslinking through the capture tag to produce aggregates or by antibody to an attached hapten. By capture tag"
herein is a meant one of a pair of binding partners as described above, such as antigen/antibody pairs, digoxygenenin, dinitrophenol, etc.

The cleaved 5' region, e.g. the detection sequence, of the signal probe, comprises a label and is detected and optionally quantitated. In one embodiment, the cleaved 5' region is hybridized to a probe on an array (capture probe) and optically detected. As described below, many signal probes can be analyzed in parallel by hybridization to their complementary probes in an array.

In a preferred embodiment, the invasive cleavage reaction is configured to utilize a fluorophore-quencher reaction. A signalling probe comprising both a fluorophore and a quencher is used, with the fluorophore and the quencher on opposite sides of the cleavage site. As will be appreciated by those in the art, these will be positioned closely together. Thus, in the absence of cleavage, very little signal is seen due to the quenching reacfion. After cleavage, however, the distance between the two is large, and thus fluorescence can be detected. Upon assembly of an assay complex, comprising the target sequence, an invader probe, and a signalling probe, and the introduction of the cleavage enzyme, the cleavage of the complex results in the disassociation of the quencher from the complex, resulting in an increase in fluorescence.

In this embodiment, suitable fluorophore-quencher pairs are as known in the art. For example, suitable quencher molecules comprise Dabcyl.

As will be appreciated by those in the art, this system can be configured in a variety of conformations, as discussed in Figure 4.

In a preferred embodiment, to obtain higher specificity and reduce the detection of contaminating uncleaved signal probe or incorrectly cleaved product, an additional enzymatic recogni6on step is introduced in the array capture procedure. For example, the cleaved signal probe binds to a capture probe to produce a double-stranded nucleic acid in the array. In this embodiment, the 3' end of the cleaved signal probe is adjacent to the 5' end of one strand of the capture probe, thereby, forming a substrate for DNA ligase (Broude et al. 1991. PNAS 91: 3072-3076). Only correctly cleaved product is ligated to the capture probe. Other incorrectly hybridized and non-cleaved signal probes are removed, for example, by heat denaturation, high stringency washes, and other methods that disrupt base pairing.

CYCLING PROBE TECHNIQUES (CPT) In a preferred embodiment, the signal amplification technique is CPT. CPT
technology is described in a number of patents and patent applications, including U.S. Patent Nos.
5,011,769, 5,403,711, 5,660,988, and 4,876,187, and PCT published applications WO 95/05480, WO
95/1416, and WO
95/00667.

Generally, CPT may be described as foilows. A CPT primer (also sometimes referred to herein as a 'sdssiie primer"), comprises two probe sequences separated by a scissile linkage. The CPT primer Is substanfialiy complementary to the target sequence and thus will hybridize to II: to form a hybridization compiex. The scissile linkage is cleaved, without cleaving the target sequence, resulting in the two probe sequences being separated. The two probe sequences can thus be more easily disassociated from the target, and the readion can be repeated any number of times. The cleaved primer is then detected as outiined herein.

By "scissile linkage' herein is meant a linkage within the sdssle probe that can be cleaved when the probe Is part of a hybridization complex, that is, when a double-stranded complex is formed. It Is important that the scissile linkage cleave only the scissile probe and not the sequence to which it is hybridized Q.e. either the target sequence or a probe sequence), such that the target sequence may be reused in the reaction for ampl'rfication of the signal. As used herein, the scissiie linkage, Is any connecting chemical structure which joins two probe sequences and which Is capable of being seiectively deaved without cleavage of eilher the probe sequences or the sequence to which the scissiie probe Is hybridized. The sdssiie linkage may be a single bond, or a multlple unit sequence.
As will be appreciated by those in the art, a number of possible sdssiie linkages may be used.

In a preferred embodiment, the scissile linkage comprises RNA. This system, previously described In as outlined above, is based on the fact that certain doubie-Wanded nucleases, partlcuiary ribonucleases, wiil nick or excise RNA nucleosides from a RNA:DNA
hybridizatlon complex. Of pari9cular use in this embodiment is RNAseH, Exo III, and reverse transcriptase.

In one embodiment, the entire scisslie probe Is made of RNA, the nicking is facilitated especially when carried out with a double-stranded ribonuciease, such as RNAseH or Exo ill.
RNA probes made entirely of RNA sequences are partlcularly useful because first, they can be more easily produced enzymatically, and second, they have more cleavage sites which are accessible to nicking or cleaving by a nicking agent, such as the ribonudeases. Thus, sdssile probes made entirely of RNA do not rely on a scissile Iinkage since the scissile linkage is Inherent In the probe.

In a preferred embodiment, when the sassile linkage is a nucleic acid such as RNA, the methods of the invention may be used to detect mismatches, as is generally described in U.S. Patent Nos. 5,660,988, and WO 95/14106. These mismatch detecfion methods are based on the fact that RNAseH may not bind to and/or deave an RNA:DNA
duplex if there are mismatches present in the sequence. Thus, in the NA,-R-NAz embodiments, IW41 and NA2 are non-RNA nucieic acids, preferably DNA. Preferably, the mismatch is within the RNA:DNA
duplex, but in some embodiments the mismatch Is present In an adjacent sequence very close to the desired sequence, close enough to affect the RNAseH (generally witbin one or two bases). Thus, In this embodiment, the nucieic acid scissile linkage is designed such that the sequence of the sdssile linkage reflects the particuiar sequence to be detected, i.e. the area of the putative mismatch.

In some embodiments of mismatch detedyon, the rate of generation of the released fragments Is such that the methods provide, essenfialiy, a yes/no resuit, whereby the detection of virtuaiiy any released fragment indicates the presence of the desired target sequence. Typically, however, when there is only a minimal mismatch (for example, a 1-, 2- or 3-base mismatch, or a 3-base deletion), there is some generafion of cleaved sequences even though the target sequence is not present. Thus, the rate of generation of cleaved fragments, and/or the finai amount of cleaved fragments, is quantified to indicate the presence or absence of the target. In addition, the use of secondary and tertiary sdssiie probes may be parbcuiariy useful in this embodiment, as this can ampiify the differences between a perfect match and a mismatch. These methods may be particuiariy useful in the determination of homozygotic or heterozygotic states of a pafiient.

In this embodiment, it is an important feature of the scissiie linkage that its length Is determined by the suspected difference between the target and the probe. In particular, this means that the scissile Unkage must be of sufficient length to encompass the suspected difference, yet short enough so that the scissile linkage cannot inappropriateiy "specifically hybridize" to the selected nucleic acid molecule when the suspected difference is present; such inappropriate hybridizafion would permit excision and thus cleavage of scissile Unkages even though the selected nucleic add molecule was not fully complementary to the nucleic acid probe. Thus In a preferred embodiment, the sdssiie linkage is between 3 to 5 nucieofides In iength, such that a suspected nucieotide difference from I nucieotide to 3 nucieotides is encompassed by the scissile linkage, and 0, 1 or 2 nucieotides are on either side of the diFference.

Thus, when the sdssiie linkage is nucleic add, preferred embodiments utilize from 1 to about 100 nucleotides, with from about 2 to about 20 being preferred and from about 5 to about 10 being particuiarly preferred.

CPT may be done enzymaticaiiy or chemically. That is, in addition to RNAseH, there are several other cleaving agents which may be useful in cleaving RNA (or other nucleic acid) scissile bonds. For example, several chemical nucleases have been reported; see for example Sigman et al., Annu. Rev.
Biochem.1990, 59, 207 236; Sigman et al., Chem. Rev. 1993,93, 2295-2316;
Bashkin et al., J. Org.
Chem. 1990, 55, 5125-5132; and Sigman et al., Nucleic Acids and Molecular Biology, vol. 3, F.
Eckstein and D.M.J. Ulley (Eds), Springer-Verlag, Heidelberg 1989, pp. 13-27.

Specific RNA hydrolysis Is also an active area; see for example Chin, Acc.
Chem. Res.1991, 24,145-152; Breslow et al., Tetrahedron, 1991, 47, 2365-2376; Anslyn et al., Angew.
Chem. lnt. Ed. Engi., 1997, 36, 432-450; and references therein. Reacave phosphate centers are also of interest in developing scissile linkages, see Hendry et al., Prog. Inorg.
Chem.:
Bioinorganic Chem. 1990, 31, 201-258.

Current approaches to site-directed RNA hydrolysis Include the conjugation of a reactlve mokity capable of cleaving phosphodiester bonds to a recogniUon element capable of sequence-speciGoally hybridizing to RNA. In most cases, a metal compiex Is covalentiy attached to a DNA stand which forms a stable heterodupiex. Upon hybridizafion, a Lewis acid Is placed. in close proximity to the RNA
backbone to effect hydrolysis; see Magda et al., J. Am. Chem. Soc.
1994,116,7439; Hall et al., Chem. Biology 1994, 1, 185-190; Bashidn et al., J. Am. Chem. Soc. 1994, 116, 5981-5982; Hall et al., Nucleic Acids Res. 1996,24,3522; Magda et al., J. Am. Chem. Soc.
1997,119,2293; and Magda et al., J. Am. Chem. Soc.1997,119, 6947.

In a similar fashion, DNA-poiyamine conjugates have been demonstrated to induce site-directed RNA
strand scission; see for example, Yoshinari et al., J. Am. Chem. Soc. 1991, 113, 5899-5901; Endo et al., J. Org. Chem. 1997, 62, 846; and Barbier et al., J. Am. Chem.
Soc.1992,114, 3511-3515.

In a preferred embodiment, the scissile 6nkege is not necessarily RNA. For example, dhemic:al cleavage moieties may be used to cleave basic sites in nucleic acids; see Belmont, et al.,New J.
Chem. 1997, 21, 47-54; and references therein. Similariy, photocleavabie moieties, for example, using transition metals, may be used; see Moucheron, et al., Inorg. Chem. 1997, 36, 584.592.

Other approaches rely on chemical mofeties or enzymes; see for example Kecdc at al., Biochemistry 1995, 34,12029-12037; Kirk et al., Chem. Commun. 1998, in press; cleavage of G-U basepaus by metal complexes; see Biochemistry,1992, 31, 5423-5429; diamine complexes for cleavage of RNA;
Komiyama, et al., J. Org. Chem. 1997,62, 2155-2160; and Chow et al., Chem.
Rev. 1997,97,1489-1513, and references therein..

The first step of the CPT method requires hybridizing a primary scissiie primer (also called a primary scissile probe) to the target. This Is preferably done at a temperature that allows both the binding of the longer primary probe and disassociation of the shorter cleaved portions of the primary probe, as will be appreciated by those in the art As oufiined herein, this may be done in soiution, or either the target or one or more of the scissile probes may be attached to a solid support. For example, it is possible to utiiize "anchor probes" on a solid support which are substantially complementary to a portion of the target sequence, preferably a sequence that is not the same sequence to which a scissile probe will bind.

Similarly, as outlined herein, a preferred embodiment has one or more of the scissite probes attached to a solid support such as a bead. In this embodiment, the soluble target diffuses to allow the formation of the hybridization complex between the soluble target sequence and the support-bound scissile probe. In this embodiment, it may be desirable to Include additional scissile linkages In the scissile probes to allow the release of two or more probe sequences, such that more than one probe sequence per scissile probe may be detected, as is ouUined below, In the Interests of maximiang the signal.

In this embodiment (and In other techniques herein), preferred methods udlize cutting or shearing techniques to cut the nucleic add sample containing the target sequence into a size that will allow sufficient diffusion of the target sequence to the surface of a bead. This may be accomplished by shearing the nucieic acid through mechanical forces (e.g. sonication) or by cieaving the nucieic add using restriction endonucleases. Altematively, a fragment containing the target may be generated using polymerase, primers and the sample as a template, as In polymerase chain reactlon (PCR). In additlon, amplification of the target using PCR or LCR or related methods may aiso be done; this may be parficulariy useful when the target sequence is present In the sample at extremely low copy numbers. Similarly, numerous techniques are known in the art to Increase the rate of mixing and hybridizafion including ag'dation, heating, techniques that increase the overall concentration such as predp'dation, drying, dialysis, centrifugation, electrophoresis, magnetic bead concentration, etc.

In general, the scissile probes are introduced In a molar excess to their targets (including both the target sequence or other scmle probes, for example when secondary or tertiary scissile probes are used), with ratios of scissile probe:target of at least about 100:1 being preferred, at least about 1000:1 being parbcularly preferred, and at least about 10,000:1 being especially preferred. In some embodiments the excess of probe:target will be much greater. In additieon, ratios such as these may be used for all the amplification techniques outlined herein.

Once the hybridization complex between the primary scissfie probe and the target has been formed, the complex Is subjected to cleavage conditions. As will be appredated, this depends on the composibon of the scissile probe; if it Is RNA, RNAseH is introduced. It should be noted that under certain circun-stances, such as is generaily outlined in WO 95/00666 and WO
96A00667, the use of a double-stranded binding agent such as RNAseH may aNow the reaction to proceed even at temperatures above the Tm of the primary probe:target hybridization complex. Accordingly, the addition of scissile probe to the target can be done either fin3t, and then the cieavage agent or cleavage conditions introduced, or the probes may be added in the presence of the cleavage agent or condifions.

The cleavage conditions result in the separation of the two (or more) probe sequences of the primary scissile probe. As a result, the shorter probe sequences will no longer remain hybridized to the target sequence, and thus the hybridization complex will disassociate, leaving the target sequence intact.
The opfimal temperature for carrying out the CPT reactions is generally from about 5 C to about 25 C
below the melting temperatures of the probe:target hybridization complex. This provides for a rapid rate of hybridization and high degree of specificity for the target sequence.
The Tm of any par6cular hybridization complex depends on salt concentration, G-C content, and length of the complex, as is known in the art and described herein.

During the reaction, as for the other amplification techniques herein, it may be necessary to suppress cleavage of the probe, as well as the target sequence, by nonspecific nucleases. Such nucleases are generally removed from the sample during the isolation of the DNA by heating or extraction procedures. A number of inhibitors of single-stranded nucleases such as vanadate, inhibitors it-ACE
and RNAsin, a placental protein, do not affect the activity of RNAseH. This may not be necessary depending on the purity of the RNAseH and/or the target sample.

These steps are repeated by allowing the reaction to proceed for a period of time. The reaction is usually carried out for about 15 minutes to about 1 hour. Generally, each molecule of the target sequence will turnover between 100 and 1000 times in this period, depending on the length and sequence of the probe, the specific reaction conditions, and the cleavage method. For example, for each copy of the target sequence present in the test sample 100 to 1000 molecules will be cleaved by RNAseH. Higher levels of amplification can be obtained by allowing the reaction to proceed longer, or using secondary, tertiary, or quaternary probes, as is outlined herein.

Upon completion of the reaction, generally determined by time or amount of cleavage, the uncleaved scissile probes must be removed or neutralized prior to detection, such that the uncleaved probe does not bind to a detection probe, causing false positive signals. This may be done in a variety of ways, as is generally described below.

In a preferred embodiment, the separation is facilitated by the use of beads containing the primary probe. Thus, when the scissile probes are attached to beads, removal of the beads by filtration, centrifugation, the application of a magnefic field, electrostatic interactions for charged beads, adhesion, etc., results in the removal of the uncleaved probes.

In a preferred embodiment, the separation is based on strong acid precipitation. This is useful to separate long (generally greater than 50 nucleotides) from smaller fragments (generally about 10 nucleotides). The introduction of a strong acid such as trichloroacetic acid into the solution causes the longer probe to precipitate, while the smaller cleaved fragments remain in solution. The solution can be centrifuged or fiRered to remove the precipitate, and the cleaved probe sequences can be quantitated.

In a preferred embodiment, the scissile probe contains both a detectable label and an affinity binding ligand or moiety, such that an af8n'ily support Is used to carry out the separafion. In this embodiment, it Is important that the detectable label used for detection Is not on the same probe sequence that contains the affinity moiety, such that removal of the uncleaved probe, and the deaved probe containing the affinity moiety, does not remove all the detectable labels.
Aitemativeiy, the sdssiie probe may contain a capture tag; the binding partner of the capture tag is attached to a soGd support such as glass beads, latex beads, dextrans, etc. and used to pull out the uncleaved probes, as Is known in the-art. The cleaved probe sequences, which do not contain the capture tag, remain in soiution and then can be detected as outtined below.

In a preferred embodiment, similar to the above embodiment, a separation sequence of nucieic acid is included In the scissiie probe, which is not cieaved during the reactlon. A
nucleic acid complementary to the separation sequence is attached to a solid support such as a bead and serves as a catcher sequence. Preferably, the separation sequence is added to the scissile probes, and Is not recognized by the target sequence, such that a generalized catcher sequence may be utllized In a variety of assays.

After removal of the uncleaved probe, as required, detection proceeds via the adcrffion of the cieaved probe sequences to the array compositlons, as outlined below. In general, the cleaved probe Is bound to a capture probe, edher diredly or indiredly, and the label is detected. in a preferred embodiment, no higher order probes are used, and detection is based on the probe sequence(s) of the primary primer. In a preferred embodiment, at least one, and preferabty more, secondary probes-(aiso- referred to herein as secondary primers) are used; the secondary probes hybridize to the domains of the cieavage probes; etc.

Thus, CPT requires, again in no pardcuiar order, a first CPT primer comprising a first probe sequence, a scissile linkage and a second probe sequence; and a cleavage agenL

In this manner, CPT results in the generation of a large amount of cieaved primers, which then can be detected as outlined below.

SANDWICH ASSAY TECHNIQUES
In a preferred embodiment, the signal ampiification technique is a"sandwich"
assay, as is generally described in U.S. Patent Nos. 5,681,702, 5,597,909, 5,545,730, 5,594,117, 5,591,584, 5,571,670, 5,580,731, 5,571,670, 5,591,584, 5,624,802, 5,635,352, 5,594,118, 5,359,100, 5,124,246 and 5,681,697. Aithough sandwich assays do not result in the alteration of primers, sandwich assays can be considered signal amplificafion techniques since multiple signals (i.e. label probes) are bound to a single target, resulfing in the amplification of the signal. Sandwich assays may be used when the target sequence does not contain a label; or when adapters are used, as outlined below.

As discussed herein, it should be noted that the sandwich assays can be used for the detection of primary target sequences (e.g. from a patient sample), or as a method to detect the product of an amplification reaction as outlined above; thus for example, any of the newly synthesized strands outlined above, for example using PCR, LCR, NASBA, SDA, etc., may be used as the "target sequence" in a sandwich assay.

As will be appreciated by those in the art, the systems of the invention may take on a large number of different configurations. In general, there are three types of systems that can be used: (1) "non-sandwich" systems (also referred to herein as "direct" detection) in which the target sequence itself is labeled with detectable labels (again, either because the primers comprise labels or due to the incorporation of labels into the newly synthesized strand); (2) systems in which label probes directly bind to the target sequences; and (3) systems in which label probes are indirectly bound to the target sequences, for example through the use of amplifier probes.

The anchoring of the target sequence to the bead is done through the use of capture probes and optionally either capture extender probes (sometimes referred to as "adapter sequences" herein).
When only capture probes are utilized, it is necessary to have unique capture probes for each target sequence; that is, the surface must be customized to contain unique capture probes; e.g. each bead comprises a different capture probe. Alternatively, capture extender probes may be used, that allow a "universal" surface, i.e. a surface containing a single type of capture probe that can be used to detect any target sequence. "Capture extender" probes have a first portion that will hybridize to all or part of the capture probe, and a second portion that will hybridize to a first por6on of the target sequence.
This then allows the generation of customized soluble probes, which as will be appreciated by those in the art is generally simpler and less costly. As shown herein, two capture extender probes may be used. This has generally been done to stabilize assay complexes for example when the target sequence is large, or when large amplifier probes (particularly branched or dendrimer amplifier probes) are used.

Detection of the amplification reactions of the invention, including the direct detecfion of amplification products and indirect detection utilizing label probes (i.e. sandwich assays), is preferably done by detecting assay complexes comprising detectable labels, which can be attached to the assay complex in a variety of ways, as is more fully described below.

Once the target sequence has preferably been anchored to the array, an amplifier probe is hybridized to the target sequence, either directly, or through the use of one or more label extender probes, which serves to allow "generic" amplifier probes to be made. As for all the steps outlined herein, this may be done simultaneously with capturing, or sequentially. Preferably, the amplifier probe contains a multiplicity of amplification sequences, although in some embodiments, as described below, the amplifier probe may contain only a single amplification sequence, or at least two amplification sequences. The amplifier probe may take on a number of different forms; either a branched conformation, a dendrimer conformation, or a linear "string" of amplification sequences. Label probes comprising detectable labels (preferably but not required to be fluorophores) then hybridize to the amplification sequences (or in some cases the label probes hybridize directly to the target sequence), and the labels detected, as is more fully outlined below.

Accordingly, the present invention provides compositions comprising an amplifier probe. By "amplifier probe" or "nucleic acid multimer" or "amplification multimer" or grammatical equivalents herein is meant a nucleic acid probe that is used to facilitate signal amplification.
Amplifier probes comprise at least a first single-stranded nucleic acid probe sequence, as defined below, and at least one single-stranded nucleic acid amplification sequence, with a multiplicity of amplificafion sequences being preferred.

Amplifier probes comprise a first probe sequence that is used, either directly or indirectly, to hybridize to the target sequence. That is, the amplifier probe itself may have a first probe sequence that is substan6ally complementary to the target sequence, or it has a first probe sequence that is substantially complementary to a portion of an additional probe, in this case called a label extender probe, that has a first portion that is substantially complementary to the target sequence. In a preferred embodiment, the first probe sequence of the amplifier probe is substantially complementary to the target sequence.

In general, as for all the probes herein, the first probe sequence is of a length sufficient to give specificity and stability. Thus generally, the probe sequences of the invention that are designed to hybridize to another nucleic acid (i.e. probe sequences, amplification sequences, portions or domains of larger probes) are at least about 5 nucleosides long, with at least about 10 being preferred and at least about 15 being especially preferred.

In a preferred embodiment, several different amplifier probes are used, each with first probe sequences that will hybridize to a different por4on of the target sequence.
That is, there is more than one level of amplification; the amplifier probe provides an amplificafion of signal due to a multiplicity of labelling events, and several different amplifier probes, each with this muitiplicity of labels, for each target sequence is used. Thus, preferred embodiments utilize at least two different pools of amplifier probes, each pool having a different probe sequence for hybridization to different portions of the target sequence; the only real limitation on the number of different amplifier probes will be the length of the original target sequence. In addition, it is also possible that the different amplifier probes contain different ampl'fication sequences, although this is generally not preferred.

In a preferred embodiment, the amplifier probe does not hybridize to the sample target sequence directly, but instead hybridizes to a first portion of a label extender probe.
This is parbcularly useful to allow the use of "generic" amplifier probes, that is, ampl'ifier probes that can be used with a variety of different targets. This may be desirable since several of the amplifier probes require special synthesis techniques. Thus, the addidon of a relatively short probe as a label extender probe Is preferred. Thus, the first probe sequence of the amplifier probe is substantially complementary to a first porflon or domain of a first label extender single-stranded nucleic acid probe. The label extender probe also contains a second portion or domain that is substantially complementary to a portion of the target sequence. Both of these portlons are preferably at least about 10 to about 50 nucleotides In length, with a range of about 15 to about 30 being preferred. The terms "first" and "second" are not meant to confer an orientation of the sequences with respect to the 5'-3' orientation of the target or probe sequences. For example, assuming a 5'-3' orientation of the complementary target sequence, the first portion may be located either 5' to the second por6on, or 3' to the second portion. For convenience herein, the order of probe sequences are generally shown from left to right.

In a preferred embodiment, more than one label extender probe-amplifier probe pair may be used, that is, n is more than 1. That is, a plurality of label extender probes may be used, each with a portion that is substantiaily complementary to a different portion of the target sequence;
this can serve as another level of amplification. Thus, a preferred embodiment utilizes pools of at least two label extender probes, with the upper limit being set by the length of the target sequence.

In a preferred embodiment, more than one label extender probe is used with a single amplifier probe to reduce non-specific binding, as is generally outfined in U.S. Patent No.
5,681,697.
In this embodiment, a first portlon of the frrst label extender probe hybridizes to a first porbon of the target sequence, and the second pordon of the first label extender probe hybridizes to a first probe sequence of the amplifier probe. A first poriion of the second label extender probe hybridizes to a second portion of the target sequence, and the second portlon of the second label extender probe hybridizes to a second probe sequence of the amplifier probe.
These form structures sometimes referred to as "cruciform" structures or configurations, and are generally done to confer stability when large branched or dendrimeric amplifier probes are used.

In addition, as will be appreciated by those in the art, the label extender probes may interact with a preamplifier probe, described below, rather than the ampfrfier probe directly.

Similarly, as outlined above, a preferred embodiment utilizes several different amplifier probes, each with first probe sequences that will hybridize to a different portion of the label extender probe. In addition, as outiined above, it is also possible that the different ampiifier probes conbain different ampiificatlon sequences, although this Is generally not preferred.

In addition to the first probe sequence, the amplifier probe aiso comprises at least one ampiificafion sequence. An'ampification sequence or'ampl'fication segment' or grammaticai equivalents herein is meant a sequence that Is used, either directiy or indirectiy, to bind to a first portlon of a label probe as is more fully described below. Preferably, the ampGfier probe comprises a muitipiidly of ampiification sequences, with from about 3 to about.1000 being preferred, from about 10 to about 100 being partlcuiariy preferred, and about 50 being espedaily preferred. In some cases, for example when linear amplifier probes are used, from 1'to about 20 is preferred with from about 5 to about 10 being parlicuiatty preferred.

The amplification sequences may be linked to each other In a variety of ways, as wUi be appredated by those in the art. They may be covaiently linked directly to each other, or to Intervening sequences or chemical moieties, through nucleic add iinkages such as phosphodiester bonds, PNA bonds, etc., or through Interposed linking agents such amino acid, carbohydrate or polyol bridges, or through other cross-iinking agents or binding partners. The site(s) of linkage may be at the ends of a segment, and/or at one or more internal nucieotides in the strand. In a preferred embodiment, the ampiification sequences are attached via nucieic acid Inkages.

In a preferred embodiment, branched ampiifier probes are used, as are generally described in U.S.
Patent No. 5,124,246, hereby Incorporated by reference. Branched amplifier probes may.take on 'fork-like" or "comb-like' conformafions. 'Fork-like" branched ampiifier probes generally have three or more oiigonucieotide segments emanating from a point of origin to form a branched structure. The point of origin may be another nucieotide segment or a muitifunctional molecule to whdh at least three segments can be covaientiy or tightiy bound. "Comb-like' branched ampiifier probes have a linear backbone with a muitipiicily of sidechain oiigonucieotides extending from the backbone. In either conformation, the pendant segments will nomaally depend from a modifled nudeotide or other organic moiety having the appropriate functionai groups for attachment of oiigonucieotides. Furthermore, in either conformation, a large number of ampiification sequences are avaiiabie for binding, efther directly or indirecdy, to detection probes. In general, these structures are made as is known in the art, using modi6ed muttifuncdonai nucleotides, as is described in U.S. Patent Nos.
5,635,352 and 5,124,246, among others.

In a preferred embodiment, dendrimer ampiifier probes are used, as are generally described In U.S.
Patent No. 5,175,270. Dendrimeric ampMer probes have ampiificatlon sequences that are attached via hybridization, and thus have portions of double-stranded nucleic acid as a component of their structure. The outer surface of the dendrimer ampiifier probe has a muitipiicity of ampiificatlon sequences.

In a preferred embodiment, linear amplifier probes are used, that have individual amplification sequences linked end-to-end either directly or with short intervening sequences to form a polymer. As with the other amplifier configurations, there may be additional sequences or moieties between the amplificafion sequences. In one embodiment, the linear amplifier probe has a single amplification sequence.

In addition, the amplifier probe may be totally linear, totally branched, totally dendrimeric, or any combination thereof.

The amplification sequences of the amplifier probe are used, either directly or indirectly, to bind to a label probe to allow detection. In a preferred embodiment, the amplification sequences of the amplifier probe are substantially complementary to a first por6on of a label probe. Alternatively, amplifier extender probes are used, that have a first portion that binds to the amplification sequence and a second poraon that binds to the first portion of the label probe.

In addition, the compositions of the invention may include "preamplifier"
molecules, which serves a bridging moiety between the label extender molecules and the amplifier probes.
In this way, more amplifier and thus more labels;are ultimately bound to the detection probes.
Preamplifier molecules may be either linear or branched, and typically contain in the range of about 30-3000 nucleotides.
Thus, label probes are either substantially complementary to an amplification sequence or to a porbon of the target sequence.

Detection of the amplification reacfions of the invention, including the direct detecfion of amplification products and indirect detecfion utilizing label probes (i.e. sandwich assays), is done by detecting assay complexes comprising labels as is outlined herein.

In addition to amplification techniques, the present invention also provides a variety of genotyping reactions that can be similarly detected and/or quantified.

GENOTYPING
In this embodiment, the invention provides compositions and methods for the detection (and optionally quanfification) of differences or variations of sequences (e.g. SNPs) using bead arrays for detection of the differences. That is, the bead array serves as a platform on which a variety of techniques may be used to elucidate the nucleofide at the position of interest ("the detection posiition"). In general, the methods described herein relate to the detection of nucleofide substitutions, although as will be appreciated by those in the art, deletions, insertions, inversions, etc. may also be detected.

These techniques fall into five general categories: (1) techniques that rely on traditional hybridization methods that utilize the variation of stringency conditions (temperature, buffer condifions, etc.) to distinguish nucleotides at the detecfion position; (2) extension techniques that add a base ("the base") to basepair with the nucleotide at the detection position; (3) ligation techniques, that rely on the specificity of ligase enzymes (or, in some cases, on the specificity of chemical techniques), such that ligafion reactions occur preferentially if perfect complementarity exists at the detection position; (4) cleavage techniques, that also rely on enzymatic or chemical specificity such that cleavage occurs preferentially if perfect complementarity exists; and (5) techniques that combine these methods.

As outlined herein, in this embodiment the target sequence comprises a position for which sequence information is desired, generally referred to herein as the "detection position" or "detection locus". In a preferred embodiment, the detection position is a single nucleotide, although in some embodiments, it may comprise a plurality of nucleotides, either contiguous with each other or separated by one or more nucleotides. By "plurality" as used herein is meant at least two. As used herein, the base which basepairs with a detection position base in a hybrid is termed a "readout position" or an "interrogation position".

In some embodiments, as is outlined herein, the target sequence may not be the sample target sequence but instead is a product of a reaction herein, somefimes referred to herein as a "secondary"
or "derivative" target sequence. Thus, for example, in SBE, the extended primer may serve as the target sequence; similarly, in invasive cleavage variations, the cleaved detecfion sequence may serve as the target sequence.

As above, if required, the target sequence is prepared using known techniques.
Once prepared, the target sequence can be used in a variety of reactions for a variety of reasons. For example, in a preferred embodiment, genotyping reactions are done. Similarly, these reactions can also be used to detect the presence or absence of a target sequence. In addition, in any reaction, quantitation of the amount of a target sequence may be done. While the discussion below focuses on genotyping reactions, the discussion applies equally to detecting the presence of target sequences and/or their quantification.

Furthermore, as outlined below for each reaction, each of these techniques may be used in a solution based assay, wherein the reaction is done in solufion and a reaction product is bound to the array for subsequent detection, or in solid phase assays, where the reaction occurs on the surface and is detected.

These reactions are generally classified into 5 basic categories, as outlined below.
SIMPLE HYBRIDIZATION GENOTYPING
In a preferred embodiment, straight hybridization methods are used to elucidate the identity of the base at the detection posi6on: Generally speaking, these techniques break down into two basic types of reactions: those that rely on compefitive hybridization techniques, and those that discriminate using stringency parameters and combinations thereof.

Comaetitive hybridization In a preferred embodiment, the use of competitive hybridization probes Is done to elucidate either the ldentity of the nucieotide(s) at the detection position or the presence of a mismatch. For example, sequencing by hybridization has been described (Drmanac et al., Genomics 4:114 (1989); Koster et al., Nature Biotechnology 14:1123 (1996); U.S. Patent Nos. 5,525,464;
5,202,231 and 5,695,940, among others.

It should be noted in this context that "mismatch' is a relative term and meant to indicate a difference in the identity of a base at a par8cular position, termed the "detection posi4on" herein, between two sequences. In general, sequences that differ from wiid type sequences are referred to as mismatches. However, par6culariy in the case of SNPs, what constitutes'wiid type.' may be difficult to determine as multiple alleles can be relativeiy frequently observed in the popuiation, and thus "mismatch' in this context requires the ardficiai adoption of one sequence as a standard. Thus, for the purposes of this invention, sequences are referred to herein as "match' and "mismatch'. Thus, the present invention may be used to detect substitutions, insertions or deietions as compared to a wild-type sequence.

In a preferred embodiment, a piuraiily of probes (sometlmes referred to herein as "readout probes") are used to Identify the base at the detection position. In this embodiment, each different readout probe comprises a different detection label (which, as outlined below, can be either a primary label or a secondary label) and a different base at the position that wili hybridize to the detection position of the target sequence (herein referred to as the readout posi6on) such that difFerential hybridiration will occur. That is, all other parameters being equal, a perfectly complementary readout probe (a "match probe") will in general be more stable and have a slower off rate than a probe comprising a mismatch (a "mismatch probe") at any par6cuiar temperature. Accordingly, by using different readout probes, each with a different base at the readout position and each with a different label, the identification of the base at the detection position is elucidated.

Accordingly, a detectable label is incorporated into the readout probe. In a preferred embodiment, a set of readout probes are used, each comprising a different base at the readout position. In some embodiments, each readout probe comprises a different label, that Is distinguishabie from the others.
For example, a first label may be used for probes comprising adenosine at the readout positlon, a second label may be used for probes comprising guanine at the readout position, etc. In a preferred embodiment, the length and sequence of each readout probe is identical except for the readout position, afthough this need not be true in all embodiments.

The number of readout probes used will vary depending on the end use of the assay. For example, many SNPs are biallelic, and thus two readout probes, each comprising an interrogation base that will basepair with one of the detection position bases. For sequencing, for example, for the discovery of SNPs, a set of four readout probes are used, although SNPs may also be discovered with fewer readout parameters.

As will be appreciated by those in the art and additionally outlined below, this system can take on a number of different configurations, including a solution phase assay and a solid phase assay.
Solution phase assay A solution phase assay that is followed by attaching the target sequence to an array is depicted in Figure 8D. In Figure 8D, a reaction with two different readout probes is shown. After the competitive hybridization has occured, the target sequence is added to the array, which may take on several configurations, outlined below.

Solid phase assay In a preferred embodiment, the competition reaction is done on the array. This system may take on several configurations.

In a preferred embodiment, a sandwich assay of sorts is used. In this embodiment, the bead comprises a capture probe that will hybridize to a first target domain of a target sequence, and the readout probe will hybridize to a second target domain, as is generally depicted in Figure 8A. In this embodiment, the first target domain may be either unique to the target, or may be an exogeneous adapter sequence added to the target sequence as outlined below, for example through the use of PCR reactions. Similarly, a sandwich assay that utilizes a capture extender probe, as described below, to attach the target sequence to the array is depicted in Figure 8C.

Alternatively, the capture probe itself can be the readout probe as is shown in Figure 8B; that is, a plurality of microspheres are used, each comprising a capture probe that has a different base at the readout position. In general, the target sequence then hybridizes preferentially to the capture probe most closely matched. In this embodiment, either the target sequence itself is labeled (for example, it may be the product of an amplification reaction) or a label probe may bind to the target sequence at a domain remote from the detection posifion. In this embodiment, since it is the locafion on the array that serves to identify the base at the detec6on position, different labels.
are not required.

In a further embodiment, the target sequence itself is attached to the array, as generally depicted for bead arrays in Figure 8E and described below.

Stringency Variation In a preferred embodiment, sensitnrily to variations in stringency parametets are used to determine either the Identity of the nucieotide(s) at the detedion position or the presence of a mismatch. As a preHminary matter, the use of different stringency conditions such as variations in temperature and buffer composi6ion to determine the presence or absence of mismatches In double stranded hybrids comprising a single stranded target sequence and a probe is well known.

With particular regard to temperature, as is known In the art, differences in the number of hydrogen bonds as a function of basepairing between perfect matches and mismatahes can be expioited as a result of their different Tms (the temperature at which 50% of the hybrid Is denatured). Accordingly, a hybrid comprising perfect complementarity will melt at a higher temperature than one comprising at least one mismatch, all other parameters being equal. (It should be noted that forthe purposes of the discussion herein, all other parameters (i.e. length of the hybrid, nature of the backbone (i.e. naturally occuting or nucieic acid analog), the assay solutlon composidon and the composition of the bases, induding G-C content are kept constant). However, as will be appredated bythose in the art, these factors may be varied as well, and then taken into account.) In general, as outlined herein, high stringency conditions are those that result in perfect matci9at remaining In hybtidization complexes, while Imperfect matches melt off.
Simiiarly, low stringency conditions are those that allow the formation of hybridization complexes with both perfect and Imperfect matches. High stringency conditlons are known in the art; see for example Maniatis et al., Molecular Cloning: A Laboratory Manual, 2d Editlon,1989, and Short Protocois In Molecular Biology, ed. Ausubel, et al.. Stringent condftu am sequence-dependent and will be different in different circumstances. Longer sequences hybridize specifically at higher temperatures. An extensive guide to the hybridization of nucleic adds is found In Tifssen. Techniques in Biochem'sky and Molecular Biology--Hybridization with Nucleic Add Probes, "Overview of principies of hybridization and the strategy of nucieic acid assays' (1993). Generally, stringent conditions are selected to be about 5-100C lower than the thermal meNing point (T,,,) for the specific sequence at a defined ionic strength pH. The T. is the temperature (under defined ionic strength, pH and nucleic acid concentration) at which 50% of the probes complementary to the target hybridize to the target sequence at equiiibrium (as the target sequences are present In excess, at T,õ
50% of the probes are occupied at equilibrium). Stringent conditions will be those In which the sait concentration is less than about 1.0 M sodium ion, typically about 0.01 to 1.0 M sodium Ion concentration (or other salts) at pH 7.0 to 8.3 and the tempen3ture is at ieast about 30'C for short probes (e.g. 10 to 50 nucieotides) and at least about 600C for long probes (e.g. greater than 50 nucieotides). Stringent conditions may also be achieved with the addi6on of destabiiizing agents such as formamide. In another embodiment, less stringent hybridization coriditions are used; for example, moderate or low stringency conditions may be used, as are known in the art;
see Maniatis and Ausubel, supra, and Tijssen, supra.

As will be appreciated by those In the art, mismatch detection using temperature may proceed In a variety of ways, and is similar to the use of readout probes as outlined above. Again, as ouitined above, a pluraiity of readout probes may be used In a sandwich format; In this embodiment, all the probes may bind at permissive, low temperatures (temperatures below the Tm of the mismatch);
however, repeating the assay at a higher temperature (above the Tm of the mismatch) oniythe perfectly matched probe may bind. Thus, this system may be run wdh readout probes with different detectable labels, as outlined above. Aitemativeiy, a single probe may be used to query whether a par6cular base is present.

Aitemativey, as described above, the capture 'probe may setve as the readout probe; in this embodiment, a single label may be used on the target; at temperatures above the Tm of the mismatch, only signals from perfect matches will be seen, as the mismatch target will melt off.
Similarly, variations in buffer composition may be used to elucidate the presence or absence of a mismatch at the detection position. Suitable conditions include, but are not limited to, formamide concentration. Thus, for example, "low" or "permissive" stringency conditions inciude formamide concentrations of 0 to 10%, while "high" or "stringent" conditions utllize formamide concentraflons of Z40%. Low stringency conditions include NaCI concentrations of Z 1 M, and high stringency conditions Include concentrations of s 0.3 M. Furthermore, low stringency condi6ons include MgClz concentrations of t 10 mM, moderate stringency as 1-10 mM, and high stringency conditlons Include concentrations of s 1 mM.

In this embodiment, as for temperature, a pluraiity of readout probes may- be used, with different bases in the readout position (and optionaiy different labels). Running the assays under the permissive conditions and repeating under stringent conditions will allow the elucidation of the base at the detection positlon.

In one embodiment, the probes used as readout probes are "Molecular Beacon"
probes as are generally described in Whitcombe et al., Nature Biotechnology 17:804 (1999), As is known in the art, Molecular Beacon probes form "haitpin" type structures, wi(h a fluorescent label on one end and a quencher on the other. In the absence of the target sequence, the ends of the hairpin hybridize, causing quenching of the label. In the presence of a target sequence, the hairpin structure is lost in favor of target sequence binding, resuldng in a loss of quenching and thus an increase In signal.

In one embodiment, the Molecular Beacon probes can be the capture probes as outlined herein for readout probes. For example, different beads comprising labeled Molecular Beacon probes (and different bases at the readout position) are made optionaliy they comprise different labels.
Aiternatively, since Molecular Beacon probes can have spectrally resolvable signals, all four probes (tf a set of four different bases with is used) differently labelled are attached to a single bead.
EXTENSION GENOTYPING
In this embodiment, any number of techniques are used to add a nucleofide to the readout position of a probe hybridized to the target sequence adjacent to the detecfion posifion.
By reiying on enzymafic specificity, preferenfially a perfectly complementary base is added. All of these methods rely on the enzymatic incorporafion of nucleofides at the detecfion posifion. This may be done using chain terminafing dNTPs, such that only a single base is incorporated (e.g. single base extension methods), or under condifions that only a single type of nucleofide is added followed by idenfification of the added nucleofide (extension and pyrosequencing techniques).

Single Base Extension In a preferred embodiment, single base extension (SBE; sometimes referred to as "minisequencing") is used to determine the idenfity of the base at the detecfion posifion. SBE
is as described above, and utilizes an extension primer that hybridizes to the target nucleic acid immediately adjacent to the detection posifion. A polymerase (generally a DNA polymerase) is used to extend the 3' end of the primer with a nucleofide analog labeled a detection label as described herein.
Based on the fidelity of the enzyme, a nucleotide is only incorporated into the readout posifion of the growing nucleic acid strand if it is perfectly complementary to the base in the target strand at the detecfion posifion. The nucleotide may be derivafized such that no further extensions can occur, so only a single nucleofide is added. Once the labeled nucleofide is added, detection of the label proceeds as outlined herein.

The reacfion is inifiated by introducing the assay complex comprising the target sequence (i.e. the array) to a solufion comprising a first nucleofide. In general, the nucleofides comprise a detectable label, which may be either a primary or a secondary label. In addition, the nucleotides may be nucleofide analogs, depending on the configurafion of the system. For example, if the dNTPs are added in sequenfial reactions, such that only a single type of dNTP can be added, the nucleofides need not be chain terminating. In addition, in this embodiment, the dNTPs may all comprise the same type of label.

Alternafively, if the reacfion comprises more than one dNTP, the dNTPs should be chain terminafing, that is, they have a blocking or protecfing group at the 3' posifion such that no further dNTPs may be added by the enzyme. As will be appreciated by those in the art, any number of nucleofide analogs may be used, as long as a polymerase enzyme will sfill incorporate the nucleotide at the readout posifion. Preferred embodiments ufilize dideoxy-triphosphate nucleofides (ddNTPs) and halogenated dNTPs. Generally, a set of nucleotides comprising ddATP, ddCTP, ddGTP and ddTTP is used, each with a different detectable label, although as outlined herein, this may not be required. Alternafive preferred embodiments use acyclo nucleotides (NEN). These chain terminafing nucleotide analogs are parficularly good substrates for Deep vent (exo ) and thermosequenase.

In addition, as will be appreciated by those in the art, the single base extension reactions of the present invention allow the precise incorporation of modified bases into a growing nucleic acid strand.
Thus, any number of modified nucleotides may be incorporated for any number of reasons, including probing structure-function relationships (e.g. DNA:DNA or DNA:protein interactions), cleaving the nucleic acid, crosslinking the nucleic acid, incorporate mismatches, etc.

As will be appreciated by those in the art, the configuration of the genotyping SBE system can take on several forms.

Solufion phase assay As for the OLA reaction described below, the reaction may be done in solution, and then the newly synthesized strands, with the base-specific detectable labels, can be detected. For example, they can be directly hybridized to capture probes that are complementary to the extension primers, and the presence of the label is then detected. This is schematically depicted in Figure 9C. As will be appreciated by those in the art, a preferred embodiment utilizes four different detectable labels, i.e.
one for each base, such that upon hybridization to the capture probe on the array, the identification of the base can be done isothermally. Thus, Figure 9C depicts the readout position 35 as not neccessarily hybridizing to the capture probe.

In a preferred embodiment, adapter sequences can be used in a solution format.
In this embodiment, a single label can be used with a set of four separate primer extension reactions. In this embodiment, the extension reaction is done in solution; each reacfion comprises a different dNTP with the label or labeled ddNTP when chain termination is desired. For each locus genotyped, a set of four different extension primers are used, each with a portion that will hybridize to the target sequence, a different readout base and each with a different adapter sequence of 15-40 bases, as is more fully outlined below. After the primer extension reaction is complete, the four separate reactions are pooled and hybridized to an array comprising complementary probes to the adapter sequences. A genotype is derived by comparing the probe intensities of the four different hybridized adapter sequences corresponding to a give locus.

In addition, since unextended primers do not comprise labels, the unextended primers need not be removed. However, they may be, if desired, as outlined below; for example, if a large excess of primers are used, there may not be sufficient signal from the extended primers competing for binding to the surface.

Alternatively, one of skill in the art could use a single label and temperature to determine the identity of the base; that is, the readout position of the extension primer hybridizes to a position on the capture probe. However, since the three mismatches will have lower Tms than the perfect match, the use of temperature could elucidate the idenfity of the detection position base.

Solid phase assay Alternatively, the reaction may be done on a surface by capturing the target sequence and then running the SBE reaction, in a sandwich type format schematically depicted in Figure 9A. In this embodiment, the capture probe hybridizes to a first domain of the target sequence (which can be endogeneous or an exogeneous adapter sequence added during an amplification reacfion), and the extension primer hybridizes to a second target domain immediately adjacent to the detecfion position.
The addifion of the enzyme and the required NTPs results in the addition of the interrogation base. In this embodiment, each NTP must have a unique label. Alternafively, each NTP
reaction may be done sequentially on a different array. As is known by one of skill in the art, ddNTP and dNTP are the preferred substrates when DNA polymerase is the added enzyme; NTP is the preferred substrate when RNA polymerase is the added enzyme.

Furthermore, as is more fully outlined below and depicted in Figure 9D, capture extender probes can be used to attach the target sequence to the bead. In this embodiment, the hybridization complex comprises the capture probe, the target sequence and the adapter sequence.

Similarly, the capture probe itself can be used as the extension probe, with its terminus being directly adjacent to the detection position. This is schematically depicted in Figure 9B. Upon the addition of the target sequence and the SBE reagents, the modified primer is formed comprising a detectable label, and then detected. Again, as for the solufion based reaction, each NTP
must have a unique label, the reactions must proceed sequentially, or different arrays must be used. Again, as is known by one of skill in the art, ddNTP and dNTP are the preferred substrates when DNA polymerase is the added enzyme; NTP is the preferred substrate when RNA polymerase is the added enzyme.

In addition, as outlined herein, the target sequence may be directly attached to the array; the extension primer hybridizes to it and the reaction proceeds.

Variations on this are shown in Figures 9E and 9F, where the the capture probe and the extension probe adjacently hybridize to the target sequence. Either before or after extension of the extension probe, a ligation step may be used to attach the capture and extension probes together for stability.
These are further described below as combination assays.

In addition, Figure 9G depicts the SBE solution reaction followed by hybridization of the product of the reaction to the bead array to capture an adapter sequence.

As will be appreciated by those in the art, the determination of the base at the detection position can proceed in several ways. In a preferred embodiment, the reaction is run with all four nucleotides (assuming all four nucleofides are required), each with a different label, as is generally outlined herein.
Alternatively, a single label is used, by using four reactions: this may be done either by using a single substrate and sequential reactions, or by using four arrays. For example, dATP
can be added to the assay complex, and the generation of a signal evaluated; the dATP can be removed and dTTP added, etc. Alternatively, four arrays can be used; the first is reacted with dATP, the second with dTTP, etc., and the presence or absence of a signal evaluated. Alternatively, the reaction includes chain terminating nucleotides such as ddNTPs or acyclo-NTPS.

Alternatively, ratiometric analysis can be done; for example, two labels, "A"
and "B", on two substrates (e.g. two arrays) can be done. In this embodiment, two sets of primer extension reactions are performed, each on two arrays, with each reaction containing a complete set of four chain terminafing NTPs. The first reaction contains two "A" labeled nucleofides and two "B"
labeled nucleofides (for example, A and C may be "A" labeled, and G and T may be "B" labeled). The second reaction also contains the two labels, but switched; for example, A and G are "A" labeled and T and C are "B"
labeled. This reaction composition allows a biallelic marker to be ratiometrically scored; that is, the intensity of the two labels in two different "color" channels on a single substrate is compared, using data from a set of two hybridized arrays. For instance, if the marker is A/G, then the first reaction on the first array is used to calculate a ratiometric genotyping score; if the marker is A/C, then the second reaction on the second array is used for the calculation; if the marker is G/T, then the second array is used, etc. This concept can be applied to all possible biallelic marker combinations. "Scoring" a genotype using a single fiber ratiometric score allows a much more robust genotyping than scoring a genotype using a comparison of absolute or normalized intensifies between two different arrays.

Removal of unextended primers In a preferred embodiment, for both SBE as well as a number of other reactions outlined herein, it is desirable to remove the unextended or unreacted primers from the assay mixture, and par6cularly from the array, as unextended primers will compete with the extended (labeled) primers in binding to capture probes, thereby diminishing the signal. The concentrafion of the unextended primers relafive to the extended primer may be relatively high, since a large excess of primer is usually required to generate efficient primer annealing. Accordingly, a number of different techniques may be used to facilitate the removal of unextended primers. As outlined above, these generally include methods based on removal of unreacted primers by binding to a solid support, protecfing the reacted primers and degrading the unextended ones, and separating the unreacted and reacted primers.

Protecfion and degradation In this embodiment, the ddTNPs or dNTPs that are added during the reaction confer protection from degradation (whether chemical or enzymatic). Thus, after the assay, the degradation components are added, and unreacted primers are degraded, leaving only the reacted primers.
Labeled protecting groups are par6cularly preferred; for example, 3'-substituted-2'-dNTPs can contain anthranylic derivatives that are fluorescent (with alkali or enzymatic treatment for removal of the protecting group).

In a preferred embodiment, the secondary label is a nuclease inhibitor, such as thiol NTPs. In this embodiment, the chain-terminating NTPs are chosen to render extended primers resistant to nucleases, such as 3'-exonucleases. Addition of an exonuclease will digest the non-extended primers leaving only the extended primers to bind to the capture probes on the array.
This may also be done with OLA, wherein the ligated probe will be protected but the unprotected liga6on probe will be digested.

In this embodiment, suitable 3'-exonucleases include, but are not limited to, exo I, exo III, exo VII, and 3'-5' exophosphodiesterases.

Alternatively, an 3' exonuclease may be added to a mixture of 3' labeled biotin/streptavidin; only the unreacted oligonucleotides will be degraded. Following exonuclease treatment, the exonuclease and the streptavidin can be degraded using a protease such as proteinase K. The surviving nucleic acids (i.e. those that were biotinylated) are then hybridized to the array.

Separation systems The use of secondary label systems (and even some primary label systems) can be used to separate unreacted and reacted probes; for example, the addition of streptavidin to a nucleic acid greatly increases its size, as well as changes its physical properties, to allow more efficient separation techniques. For example, the mixtures can be size fractionated by exclusion chromatography, affinity chromatography, filtrafion or differential precipitation.
Non-terminated extension In a preferred embodiment, methods of adding a single base are used that do not rely on chain termination. That is, similar to SBE, enzymatic reacfions that utilize dNTPs and polymerases can be used; however, rather than use chain terminating dNTPs, regular dNTPs are used. This method relies on a time-resolved basis of detection; only one type of base is added during the reaction. Thus, for example, four different reactions each containing one of the dNTPs can be done; this is generally accomplished by using four different substrates, although as will be appreciated by those in the art, not all four reactions need occur to identify the nucleo6de at a detection posifion. In this embodiment, the signals from single additions can be compared to those from muitiple additions; that is, the addition of a single ATP can be distinguished on the basis of signal intensity from the addition of two or three ATPs. These reactions are accomplished as outlined above for SBE, using extension primers and polymerases; again, one label or four different labels can be used, although as outlined herein, the different NTPs must be added sequentially.

A preferred method of extension in this embodiment is pyrosequencing.
PYrosequencinq Pyrosequencing is an extension and sequencing method that can be used to add one or more nucieo6des to the detection position(s); it is very similar to SBE except that chain terminating NTPs need not be used (aithough they may be). Pyrosequencing relies on the detectlon of a reaotlon product, PPi, produced during the addition of an NTP to a growing oligonucleotide chain, rather than on a label attached to the nucieotide. One molecule of PPI Is produced per dNTP added to the extension primer. That is, by running sequential reactions with each of the nucieotides, and monitoring the reaction products, the identity of the added base Is determined.

The release of pyrophosphate (PPi) during the DNA polymerase reactlon can be quantdativeiy measured by many different methods and a number of enzymaflc methods have been described; see Reeves et al., Anal. Biochem. 28282 (1969); Guillory et al., Anal. Biochem.
39:170 (1971); Johnson et al., Anal. Biochem.15:273.(1968); Cook et al., Anal. Biochem. 91:557 (1978);
Drake et al., Anal.
Biochem. 94:117 (1979); W093/23564; WO 9828440; W098/1 3523; Nyren et ai., Anal. Biochem.
151:504 (1985); The tatter method slbws continuous monitoring of PPi and has been termed EUDA (Enzymatic Luminometrk: inorga-dc Pyrophosphate Detection Assay). A preferred embodiment u6iizes any method which can resutt In the generation of an opticai signal, with preferred embodiments utiiizing the generation of a chemiluminescent or fluorescent signal.

A preferred method monitors the creation of PPI by the conversion of PPi to ATP by the enzyme suifuryiase, and the subsequent production of visible light by flrefly luciferase (see Ronaghi et al., Science 281:363 (1998), incorporated by reference). In this method, the four deoxynucleotides (dATP, dGTP, dCTP and dTTP; coiiectiveiy dNTPs) are added stepvrise to a pardai duplex comprising a sequencing primer hybridized to a single stranded DNA template and incubated with DNA polymerase, ATP sulfurylase, luciferase, and optionaiiy a nucieotide-degrading enzyme such as apyrase. A dNTP
is only incorporated into the growing DNA strand if it Is complementary to the base In the template strand. The synthesis of DNA is accompanied by the release of PPi equal in molarity to the incorporated dNTP. The PPi is converted to ATP and the light generated by the iuciferase is directly proportionai to the amount of ATP. In some cases the unincorporated dNTPs and the produced ATP
are degraded between each cycle by the nucieotide degrading enzyme.

Accordingly, a preferred embodiment of the methods of the invention is as follows. A substrate comprising microspheres containing the target sequences and extension primers, forming hybridization complexes, is dipped or contacted with a reaction volume (chamber or well) comprising a single type of dNTP, an extension enzyme, and the reagents and enzymes necessary to detect PPi. If the dNTP Is complementary to the base of the target portion of the target sequence adjacent to the extension primer, the dNTP is added, releasing PPi and generating detectable light, which is detected as generally described in U.S.S.N.s 09/151,877 and 09/189,543, and PCT
US98/09163.
If the dNTP Is not complementary, no detectable signal resuits.

The substrate is then contacted with a second reaction volume (chamber) comprising a different dNTP
and the additional components of the assay. This process is repeated if the identity of a base at a second detectlon position is desirabie.

In a preferred embodiment, washing steps, i.e. the use of washing chambers, may be done in between the dNTP reactlon chambers, as required. These washing chambers may optionally comprise a nucleotide-degrading enzyme, to remove any unreacted dNTP and decreasing the background signal, as is described in WO 9828440.

As will be appreciated by those in the art, the system can be configured in a variety of ways, Including both a linear progression or a circular one; for example, four arrays may be used that each can dip Into one of four reaction chambers arrayed in a circular pattern. Each cyde of sequendng and reading is followed by a 90 degree rotation, so that each substrate then dips Into the next reaction well.

In a preferred embodiment, one or more Internal control sequences are used.
That is, at least one microsphere in the array comprises a known sequence that can be used to verify that the reactions are proceeding correctiy. In a preferred embodiment, at least four control sequences are used, each of which has a different nucleotide at each posiGon: the first control sequence wili have an adenosine at position 1, the second will have a cytosine, the third a guanosine, and the fourth a thymidine, thus ensuring that at least one control sequence is "lighting up' at each step to serve as an intemal control.
As for simple extension and SBE, the pyrosequencing systems may be configured In a variety of ways;
for example, the target sequence may be attached to the bead in a variety of ways, inciuding direct attachment of the target sequence; the use of a capture probe with a separate extension probe; the use of a capture extender probe, a capture probe and a separate extension probe; the use of adapter sequences In the target sequence with capture and extension probes; and the use of a capture probe that also serves as the extension probe.

One additional benefd of pyrosequencing for genotyping purposes is that since the reactlon does not rely on the incorporation of.labels into a growing chain, the unreacted extension primers need not be removed.

Allelic PCR
In a preferred embodiment, the method used to detect the base at the detectlon positlon Is alielic PCR, referred to herein as "aPCR". As described in Newton et al., Nucl. Acid Res.1T2503 (1989), hereby expressly incoporated by reference, allelic PCR allows single base discrimination based on the fact that the PCR reaction does not proceed well if the terminal 3'-nucleotide is mismatched, assuming the DNA polymerase being used lacks a 3'-exonuclease proofreading activity.
Accordingly, the identification of the base proceeds by using alielic PCR primers (sometimes referred to herein as aPCR primers) that have readout positions at their 3' ends. Thus the target sequence comprises a first domain comprising at its 5' end a detection position.

In general, aPCR may be briefly described as follows. A double stranded target nucleic acid is denatured, generally by raising the temperature, and then cooled in the presence of an excess of a aPCR primer, which then hybridizes to the first target strand. If the readout position of the aPCR
primer basepairs correctly with the detection position of the target sequence, a DNA polymerase (again, that lacks 3'-exonuclease activity) then acts to extend the primer with dNTPs, resulfing in the synthesis of a new strand forming a hybridization complex. The sample is then heated again, to disassociate the hybridization complex, and the process is repeated. By using a second PCR primer for the complementary target strand, rapid and exponential amplificafion occurs. Thus aPCR steps are denaturation, annealing and extension. The paraculars of aPCR are well known, and include the use of a thermostabie polymerase such as Taq I polymerase and thermal cycling.

Accordingly, the aPCR reaction requires at least one aPCR primer, a polymerase, and a set of dNTPs.
As outlined herein, the primers may comprise the label, or one or more of the dNTPs may comprise a label.

Furthermore, the aPCR reaction may be run as a competition assay of sorts. For example, for biallelic SNPs, a first aPCR primer comprising a first base at the readout position and a first label, and a second aPCR primer comprising a different base at the readout posifion and a second label, may be used. The PCR primer for the other strand is the same. The examination of the ratio of the two colors can serve to identify the base at the detection position.

In general, as is more fully outlined below, the capture probes on the beads of the array are designed to be substantially complementary to the extended part of the primer; that is, unextended primers will not bind to the capture probes.

LIGATION TECHNIQUES FOR GENOTYPING
In this embodiment, the readout of the base at the detection position proceeds using a ligase. In this embodiment, it is the specificity of the ligase which is the basis of the genotyping; that is, ligases generally require that the 5' and 3' ends of the ligafion probes have perfect complementarity to the target for ligation to occur. Thus, in a preferred embodiment, the identity of the base at the detection position proceeds utilizing OLA as described above, as is generally depicted in Figure 10. The method can be run at least two different ways; in a first embodiment, only one strand of a target sequence is used as a template for ligafion; alternatively, both strands may be used; the latter is generally referred to as Ligation Chain Reaction or LCR.

This method is based on the fact that two probes can be preferentially ligated together, if they are hybridized to a target strand and if perfect complementarity exists at the two bases being ligated together. Thus, in this embodiment, the target sequence comprises a configuous first target domain comprising the detecfion posifion and a second target domain adjacent to the detecfion posifion. That is, the detecfion posifion is "between" the rest of the first target domain and the second target domain.
A first ligation probe is hybridized to the first target domain and a second ligafion probe is hybridized to the second target domain. If the first ligafion probe has a base perfectly complementary to the detecfion posifion base, and the adjacent base on the second probe has perfect complementarity to its posifion, a ligafion structure is formed such that the two probes can be ligated together to form a ligated probe. If this complementarity does not exist, no ligation structure is formed and the probes are not ligated together to an appreciable degree. This may be done using heat cycling, to allow the ligated probe to be denatured off the target sequence such that it may serve as a template for further reactions. In addifion, as is more fully outlined below, this method may also be done using ligation probes that are separated by one or more nucleofides, if dNTPs and a polymerase are added (this is sometimes referred to as "Genetic Bit" analysis).

In a preferred embodiment, LCR is done for two strands of a double-stranded target sequence. The target sequence is denatured, and two sets of probes are added: one set as outlined above for one strand of the target, and a separate set (i.e. third and fourth primer probe nucleic acids) for the other strand of the target. In a preferred embodiment, the first and third probes will hybridize, and the second and fourth probes will hybridize, such that amplificafion can occur.
That is, when the first and second probes have been attached, the ligated probe can now be used as a template, in addition to the second target sequence, for the attachment of the third and fourth probes.
Similarly, the ligated third and fourth probes will serve as a template for the attachment of the first and second probes, in addition to the first target strand. In this way, an exponenfial, rather than just a linear, amplificafion can occur.

As will be appreciated by those in the art, the ligafion product can be detected in a variety of ways.
Preferably, detection is accomplished by removing the unligated labeled probe from the reacfion before applicafion to a capture probe. In one embodiment, the unligated probes are removed by digesting 3' non-protected oligonucleofides with a 3' exonuclease, such as, exonuclease I. The ligafion products are protected from exo I digesfion by including, for example, the use of a number of sequenfial phosphorothioate residues at their 3' terminus (for example at least four), thereby, rendering them resistant to exonuclease digesfion. The unligated detecfion oligonucleofides are not protected and are digested.

As for most or all of the methods described herein, the assay can take on a solufion-based form or a solid-phase form.

Solution based OLA

In a preferred embodiment, as shown in Figure 10A, the ligation reacfion is run in solution. In this embodiment, only one of the primers carries a detectable label, e.g. the first ligation probe, and the capture probe on the bead is substantially complementary to the other probe, e.g. the second ligation probe. In this way, unextended labeled ligation primers will not interfere with the assay. This substantially reduces or eliminates false signal generated by the optically-labeled 3' primers.
In addition, a solution-based OLA assay that utilizes adapter sequences may be done. In this embodiment, rather than have the target sequence comprise the adapter sequences, one of the ligafion probes comprises the adapter sequence. This facilitates the creation of "universal arrays".
For example, as depicted in Figure 10E, the first ligation probe has an adapter sequence that is used.
to attach the ligated probe to the array.

Again, as outlined above for SBE, unreacted ligafion primers may be removed from the mixture as needed. For example, the first ligation probe may comprise the label (either a primary or secondary label) and the second may be blocked at its 3' end with an exonuclease blocking moiety; after ligation and the introduction of the nuclease, the labeled ligation probe will be digested, leaving the ligation product and the second probe; however, since the second probe is unlabeled, it is effectively silent in the assay. Similarly, the second probe may comprise a binding partner used to pull out the ligated probes, leaving unligated labeled ligation probes behind. The binding pair is then disassociated and added to the array.

Solid phase based OLA
Alternatively, the target nucleic acid is immobilized on a solid-phase surface. The OLA assay is performed and unligated oligonucleotides are removed by washing under appropriate stringency to remove unligated oligonucleotides and thus the label. For example, as depicted in Figure 10B, the capture probe can comprise one of the ligation probes. Similarly, Figures 10C
and 10D depict alternative attachments.

Again, as outlined above, the detection of the OLA reacfion can also occur directly, in the case where one or both of the primers comprises at least one detectable label, or indirectly, using sandwich assays, through the use of additional probes; that is, the ligated probes can serve as target sequences, and detection may utilize amplification probes, capture probes, capture extender probes, label probes, and label extender probes, etc.

Solid Phase Oliponucleotide Ligafion Assay (SPOLA) In a preferred embodiment, a novel method of OLA is used, termed herein "solid phase oligonucleotide assay", or "SPOLA". In this embodiment, the ligation probes are both attached to the same site on the surface of the array (e.g. when microsphere arrays are used, to the same bead), one at its 5' end (the "upstream probe") and one at its 3' end (the "downstream probe"), as is generally depicted in Figure 11. This may be done as Is will be appredated by those in the art. At least one of the probes is attached via a cleavable rmker, that upon cleavage, forms a reactive or detectable (fluorophore) moiety. If ligation occurs, the read3ve moiety remains associated wlth the surface; but if no ligation occurs, due to a mismatch, the reactive moiety is free in solution to diffuse away from the surface of the array. The reactive moiety Is then used to add a detectable label.

Generally, as will be appreciated by those in the art, deavage of the cleavable finker should result In asymmetrical products; i.e. one of the "ends" should be reacfive, and the other should not, with the conflguration of the system such that the reacfive moiety remains associated with the surface if ligation occurred. Thus, for example, amino acids or succinate esters can be deaved eidter enzymatkaily (via peptidases (aminopeptidase and carboxypeptidase) or proteases) or chemically (add/base hydrolysis) to produce an amine and a carboxyl group. One of these groups can then be used to add a detectable label, as will be appreciated by those In the art and discussed herein.

Padlock probe Iloation In a preferred embodiment, the ligation probes are speciaiized probes called "padlock probes".
Nilsson et al, 1994, Science 265:2085. These probes heve a fitst ligation domain that is identical to a first ligation probe, in that it hybridizes to a first target sequence domain, and a second ligation domain, idenficai to the second ligation probe, that hybridizes to an adjacent target sequence domain. Again, as for OLA, the detection positlon can be either at the 3' end of the first ligation domain or at the 5' end of the second ligation domain.
However, the iwo ligation domains are connected by a linker, frequentiy nucieic acid. The conflguration of the system is such that upon iigation of the first and second Nga6on domains of the padlock probe, the probe forms a dreular probe, and forms a complex with the target sequence wherein the target sequence is "inserted" into the loop of the circle.

In this embodiment, the unligated probes may be removed through degradation (for example, through a nuclease), as there are no "free ends" in the ligated probe.

CLEAVAGE TECHNIQUES FOR GENOTYPING
In a preferred embodiment, the spedficity for genotyping is provided by a cleavage enzyme. There are a variety of enzymes known to cleave at specific sites, either based on sequence spedfldty, such as restricdon endonucleases, or using structurai specificity, such as is done through the use of invasive deavage technology.
ENDONUCLEASE TECHNIQUES
In a preferred embodiment, enzymes that rely on sequence specificity are used.
In general, these systems rely on the cleavage of double stranded sequence containing a spedfic sequence recognized by a nuciease, preferably an endonuclease inciuding resolvases.

These systems may work in a variety of ways, as is generally depicted in Figure 12. In one embodiment (Figure 12A), a labeled readout probe (generally attached to a bead of the array) is used;
the binding of the target sequence forms a double stranded sequence that a restricfion endonuclease can then recognize and cleave, if the correct sequence is present. An enzyme resuifing in "sficky ends" is shown in Figure 12A. The cleavage results in the loss of the label, and thus a loss of signal.
Alternatively, as will be appreciated by those in the art, a labelled target sequence may be used as well; for example, a labelled primer may be used in the PCR amplificafion of the target, such that the label is incorporated in such a manner as to be cleaved off by the enzyme.

Alternatively, the readout probe (or, again, the target sequence) may comprise both a fluorescent label and a quencher, as is known in the art and depicted in Figure 12B. In this embodiment, the label and the quencher are attached to different nucleosides, yet are close enough that the quencher molecule results in little or no signal being present. Upon the introducfion of the enzyme, the quencher is cleaved off, leaving the label, and allowing signalling by the label.

In addition, as will be appreciated by those in the art, these systems can be both solufion-based assays or solid-phase assays, as outlined herein.

Furthermore, there are some systems that do not require cleavage for detecfion; for example, some nucleic acid binding proteins will bind to specific sequences and can thus serve as a secondary label.
For example, some transcripfion factors will bind in a highly sequence dependent manner, and can disfinguish between two SNPs. Having bound to the hybridization complex, a detectable binding partner can be added for detecfion. In addifion, mismatch binding proteins based on mutated transcription factors can be used.

In addifion, as will be appreciated by those in the art, this type of approach works with other cleavage methods as well, for example the use of invasive cleavage methods, as outlined below.

Invasive cleavage In a preferred embodiment, the determinafion of the idenfity of the base at the detection posifion of the target sequence proceeds using invasive cleavage technology. As outlined above for amplificafion, invasive cleavage techniques rely on the use of structure-specific nucleases, where the structure can be formed as a result of the presence or absence of a mismatch. Generally, invasive cleavage technology may be described as follows. A target nucleic acid is recognized by two disfinct probes. A
first probe, generally referred to herein as an "invader" probe, is substanfially complementary to a first por6on of the target nucleic acid. A second probe, generally referred to herein as a "signal probe", is partially complementary to the target nucleic acid; the 3' end of the signal oligonucleofide is substanfially complementary to the target sequence while the 5' end is non-complementary and preferably forms a single-stranded "tail" or "arm". The non-complementary end of the second probe preferably comprises a "generic" or "unique" sequence, frequently referred to herein as a"detection sequence", that is used to indicate the presence or absence of the target nucleic acid, as described below. The detection sequence of the second probe preferably comprises at least one detectable label. Alternative methods have the detection sequence functioning as a target sequence for a capture probe, and thus rely on sandwich configurations using label probes.

Hybridization of the first and second oligonucleotides near or adjacent to one another on the target nucleic acid forms a number of structures. In a preferred embodiment, a forked cleavage structure, as shown in Figure 13, forms and is a substrate of a nuclease which cleaves the detection sequence from the signal oligonucleotide. The site of cleavage is controlled by the distance or overlap between the 3' end of the invader oligonucleotide and the downstream fork of the signal oligonucleo6de. Therefore, neither oligonucleotide is subject to cleavage when misaligned or when unattached to target nucleic acid.

As above, the invasive cleavage assay is preferably performed on an array format. In a preferred embodiment, the signal probe has a detectable label, attached 5' from the site of nuclease cleavage (e.g. within the detection sequence) and a capture tag, as described herein for removal of the unreacted products (e.g. biotin or other hapten) 3' from the site of nuclease cleavage. After the assay is carried out, the uncleaved probe and the 3' portion of the cleaved signal probe (e.g. the the detection sequence) may be extracted, for example, by binding to streptavidin beads or by crosslinking through the capture tag to produce aggregates or by antibody to an attached hapten. By "capture tag"
herein is a meant one of a pair of binding partners as described above, such as antigen/antibody pairs, digoxygenenin, dinitrophenol, etc.

The cleaved 5' region, e.g. the detection sequence, of the signal probe, comprises a label and is detected and optionally quantitated. In one embodiment, the cleaved 5' region is hybridized to a probe on an array (capture probe) and optically detected (Figure 13). As described below, many different signal probes can be analyzed in parallel by hybridization to their complementary probes in an array.
In a preferred embodiment as depicted in Figure 13, combination techniques are used to obtain higher specificity and reduce the detection of contaminating uncleaved signal probe or incorrectly cleaved product, an enzymatic recognition step is introduced in the array capture procedure. For example, as more fully outlined below, the cleaved signal probe binds to a capture probe to produce a double-stranded nucleic acid in the array. In this embodiment, the 3' end of the cleaved signal probe is adjacent to the 5' end of one strand of the capture probe, thereby, forming a substrate for DNA ligase (Broude et al. 1991. PNAS 91: 3072-3076). Only correctly cleaved product is ligated to the capture probe. Other incorrectly hybridized and non-cleaved signal probes are removed, for example, by heat denaturation, high stringency washes, and other methods that disrupt base pairing.

Accordingly, the present invention provides methods of determining the identity of a base at the detection position of a target sequence. in this embodiment, the target sequence comprises, 5' to 3', a first target domain comprising an overiap domain comprising at least a nucleotide in the detecbon position, and a second target domain con6guous with the detection posi6on. A
first probe (the "invader probe') is hybridized to the first target domain of the target sequence. A
second probe (the `signai probel, comprising a first porBon that hybridizes to the second target domain of the target sequence and a second portion that does not hybridize to the target sequence, Is hybridized to the second target domain. If the second probe comprises a base that is perfectiyy complementary to the detectlon position a cleavage structure Is formed. The addition of a cleavage enzyme, such as Is described In U.S. Patent Nos. 5,846,717; 5,614,402; 5,7191'029; 5,541,311 and 5,843,669, resuits In the cleavage of the detecdon sequence from the signalling probe. This then can be used as a target sequence in an assay complex.

In addition, as for a variety of the techniques outlined herein, unreacted probes (i.e. signalling probes, in the case of invasive cleavage), may be removed using any number of techniques. For example, the use of a binding partner (70 in Figure 13C) coupled-to a solid support comprising the other member of the binding pair can be done. Similarly, after cleavage of the primary signal probe, the newly created cleavage products can be seiectively labeled at the 3' or 5' ends using enzymatic or chemical methods.

Again, as outlined above, the detection of the invasive cleavage reaction can occur directly, in the case where the detec6on sequence comprises at least one label, or indirectiy, using sandwich assays, through the use of additional probes; that is, the detection sequences can serve as target sequences, and detection may utilize ampiification probes, capture probes, capture extender probes, label probes, and label extender probes, etc.

In addiGon, as for most of the techniques oufiined herein, these techniques may be done for the two strands of a double-stranded target sequence. The target sequence is denatured, and two sets of probes are added: one set as outlined above for one strand of the target, and a separate set for the other strand of the target.

Thus, the invasive cleavage reaction requires, in no par8cular order, an Invader probe, a signalling probe, and a cleavage enzyme.

As for other methods outiined herein, the invasive cleavage reaction may be done as a solution based assay or a solid phase assay.

Solution-based invasive cleavage The invasive cleavage reaction may be done in soiution, followed by addition of one of the components to an array, with optional (but preferable) removal of unreacted probes. For example, as depicted in Figure 13C, the reacfion is carried out in solution, using a capture tag (i.e. a member of a binding partner pair) that is separated from the label on the detection sequence with the cleavage site.
After cleavage (dependent on the base at the detection position), the signalling probe is cleaved. The capture tag is used to remove the uncleaved probes (for example, using magnetic particies comprising the other member of the binding pair), and the remaining solu6on is added to the array. Figure 13C
depicts the direct attachment of the detection sequence to the capture probe.
In this embodiment, the detecfion sequence can effectively act as an adapter sequence. In alternate embodiments, as depicted in Figure 13D, the detection sequence is unlabelled and an additional label probe is used; as outlined below, this can be ligated to the hybridization complex.
Solid-phase based assays The invasive cleavage reaction can also be done as a solid-phase assay. As depicted in Figure 13A, the target sequence can be attached to the array using a capture probe (in addition, although not shown, the target sequence may be directly attached to the array). In a preferred embodiment, the signalling probe comprises both a fluorophore label (attached to the pordon of the signalling probe that hybridizes to the target) and a quencher (generally on the detection sequence), with a cleavage site in between. Thus, in the absence of cleavage, very little signal is seen due to the quenching reacfion.
After cleavage, however, the detection sequence is removed, along with the quencher, leaving the unquenched fluorophore. Similarly, the invasive probe may be attached to the array, as depicted in Figure 13B.

In a preferred embodiment, the invasive cleavage reaction is configured to utilize a fluorophore-quencher reaction. A signalling probe comprising both a fluorophore and a quencher is attached to the bead. The fluorophore is contained on the portion of the signalling probe that hybridizes to the target sequence, and the quencher is contained on a porbon of the signalling probe that is on the other side of the cleavage site (termed the "detection sequence" herein). In a preferred embodiment, it is the 3' end of the signalling probe that is attached to the bead (although as will be appreciated by those in the art, the system can be configured in a variety of different ways, including methods that would result in a loss of signal upon cleavage). Thus, the quencher molecule is located 5' to the cleavage site. Upon assembly of an assay complex, comprising the target sequence, an invader probe, and a signalling probe, and the introducfion of the cleavage enzyme, the cleavage of the complex results in the disassociafion of the quencher from the complex, resulting in an increase in fluorescence.

In this embodiment, suitable fluorophore-quencher pairs are as known in the art. For example, suitable quencher molecules comprise Dabcyl.

COMBINATION TECHNIQUES
It is also possible to combine two or more of these techniques to do genotyping, quantification, detecfion of sequences, etc.

Novel combinafion of competitive hybridizafion and extension In a preferred embodiment, a combination of competitive hybridization and extension, particularly SBE, is used. This may be generally described as follows. In this embodiment, different extension primers comprising different bases at the readout position are used. These are hybridized to a target sequence under stringency conditions that favor perfect matches, and then an extension reaction is done. Basically, the readout probe that has the match at the readout position will be preferentially extended for two reasons; first, the readout probe will hybridize more efficiently to the target (e.g. has a slower off rate), and the extension enzyme will preferentially add a nucleotide to a "hybridized" base.
The reactions can then be detected in a number of ways, as outlined herein.

The system can take on a number of configurations, depending on the number of labels used, the use of adapters, whether a solution-based or surface-based assay is done, etc.
Several preferred embodiments are shown in Figure 14.

In a preferred embodiment, at least two different readout probes are used, each with a different base at the readout position and each with a unique detectable label that allows the identification of the base at the readout position. As described herein, these detectable labels may be either primary or secondary labels, with primary labels being preferred. As for all the competitive hybridization reactions, a compefition for hybridization exists with the reaction conditions being set to favor match over mismatch. When the correct match occurs, the 3' end of the hybridization complex is now double stranded and thus serves as a template for an extension enzyme to add at least one base to the probe, at a posifion adjacent to the readout position. As will be appreciated by those in the art, for most SNP analysis, the nucleotide next to the detection position will be the same in all the reac6ons.
In one embodiment, chain terminating nucleotides may be used; alternatively, non-terminafing nucleotides may be used and muitiple nucleotides may be added, if desired. The latter may be par6cularly preferred as an amplificafion step of sorts; if the nucleotides are labelled, the addition of multiple labels can result in signal amplification.

In a preferred embodiment, the nucleofides are analogs that allow separation of reacted and unreacted primers as described herein; for example, this may be done by using a nuclease blocking moiety to protect extended primers and allow preferentially degradation of unextended primers or biotin (or iminobiotin) to preferentially remove the extended primers (this is done in a solution based assay, followed by elution and addition to the array).

As for the other reactions outlined herein, this may be done as a solution based assay, or a solid phase assay. Solution based assays are generally depicted in Figures 14A, 14B
and 14C. In a solid phase reaction, an example of which is depicted in Figure 14D, the capture probe serves as the readout probe; in this embodiment, different positions on the array (e.g.
different beads) comprise different readout probes. That is, at least two different capture/readout probes are used, with three and four also possible, depending on the aliele. The reaction is run under conditions that favor the formation of perfect match hybridization complexes. In this embodiment, the dNTPs comprise a detectable label, preferably a primary label such as a fluorophore. Since the competitive readout probes are spatially defined in the array, one fluorescent label can distinguish between the alleles;
furthermore, it is the same nucleotide that is being added in the reaction, since it is the position adjacent to the SNP that is being extended. As for all the competitive assays, relative fluorescence intensity disfinguishes between the alleles and between homozygosity and heterozygosity. In addition, muitiple extension reactions can be done to amplify the signal.

For both solution and solid phase reactions, adapters may be additionally used. In a preferred embodiment, as shown in Figure 14B for the solufion based assay (although as will be appreciated by those in the art, a solid phase reaction may be done as well), adapters on the 5' ends of the readout probes are used, with identical adapters used for each allele. Each readout probe has a unique detectable label that allows the determination of the base at the readout position. After hybridization and extension, the readout probes are added to the array; the adapter sequences direct the probes to paracular array locations, and the relative intensities of the two labels distinguishes between alleles.
Alternatively, as depicted in Figure 14C for the solution based assay (although as will be appreciated by those in the art, a solid phase reaction may be done as well), a different adapter may be used for each readout probe. In this embodiment, a single label may be used, since spatial resolution is used to distinguish the alleles by having a unique adapter attached to each allelic probe. After hybridization and extension, the readout probes are added to the array; the unique adapter sequences direct the probes to unique array locations. In this embodiment, it is the relative intensities of two array positions that distinguishes between alleles.

As will be appreciated by those in the art, any array may be used in this novel method, including both ordered and random arrays. In a preferred embodiment, the arrays may be made through spotting techniques, photolithographic techniques, printing techniques, or preferably are bead arrays.

Combination of competitive hybridization and invasive cleavaQe In a preferred embodiment, a combinafion of competitive hybridization and invasive cleavage is done.
As will be appreciated by those in the art, this technique is invasive cleavage as described above, with at least two sets of probes comprising different bases in the readout position. By running the reactions under conditions that favor hybridization complexes with perfect matches, different alleles may be distinguished.

In a preferred embodiment, this technique is done on bead arrays.

Novel combination of invasive cleavage and ligation In a preferred embodiment, invasive cleavage and ligation is done, as is generally depicted in Figure 15. In this embodiment, the specificity of the invasive cleavage reaction is used to detect the nucleotide in the detection position, and the specificity of the ligase reaction is used to ensure that only cleaved probes give a signal; that is, the ligation reaction confers an extra level of specificity.

The detection sequence, comprising a detectable label, of the signal probe is cleaved if the correct basepairing is present, as outlined above. The detection sequence then serves as the "target sequence" in a secondary reaction for detection; it is added to a capture probe on a microsphere. The capture probe in this case comprises a first double stranded portion and a second single stranded portion that will hybridize to the detection sequence. Again, preferred embodiments utilize adjacent portions, although dNTPs and a polymerase to fill in the "gap" may also be done. A ligase is then added. As shown in Figure 15A, only if the signal probe has been cleaved will ligation occur; this results in covalent attachment of the signal probe to the array. This may be detected as outlined herein; preferred embodiments utilize stringency conditions that will discriminate between the ligated and unligated systems.

As will be appreciated by those in the art, this system may take on a number of configurations, including solution based and solid based assays. In a preferred embodiment, as outlined above, the system is configured such that only if cleavage occurs will ligation happen.
In a preferred embodiment, this may be done using blocking moieties; the technique can generally be described as follows. An invasive cleavage reaction is done, using a signalling probe that is blocked at the 3' end.
Following cleavage, which creates a free 3' terminus, a ligafion reaction is done, generally using a template target and a second ligation probe comprising a detectable label.
Since the signalling probe has a blocked 3' end, only those probes undergoing cleavage get ligated and labelled.

Alternatively, the orientations may be switched; in this embodiment, a free 5' phosphate is generated and is available for labeling.

Accordingly, in this embodiment, a solution invasive cleavage reaction is done (although as will be appreciated by those in the art, a support bound invasive cleavage reaction may be done as well).
As will be appreciated by those in the art, any array may be used in this novel method, including both ordered (predefined) and random arrays. In a preferred embodiment, the arrays may be made through spotting techniques, photolithographic techniques, printing techniques, or preferably are bead arrays.

Combination of invasive cleavacge and extension In a preferred embodiment, a combinafion of invasive cleavage and extension reactions are done, as generally depicted in Figure 16. The technique can generally be described as follows. An invasive cleavage reaction is done, using a signalling probe that is blocked at the 3' end. Following cleavage, which creates a free 3' terminus, an extension reaction is done (either enzymatically or chemically) to add a detectable label. Since the signalling probe has a blocked 3' end, only those probes undergoing cleavage get labelled.

Alternatively, the orientations may be switched, for example when chemical extension or labeling is done. In this embodiment, a free 5' phosphate is generated and is available for labeling.

In a preferred embodiment, the invasive cleavage reaction is configured as shown in Figure 16B. In this embodiment, the signalling probe is attached to the array at the 5' end (e.g. to the detection sequence) and comprises a blocking moiety at the 3' end. The blocking moiety serves to prevent any alteration (including either enzymatic alteration or chemical alteration) of the 3' end. Suitable blocking moieties include, but are not limited to, chain terminators, alkyl groups, halogens; basically any non-hydroxy moiety.

Upon formation of the assay complex comprising the target sequence, the invader probe, and the signalling probe, and the introduction of the cleavage enzyme, the poraon of the signalling probe comprising the blocking moiety is removed. As a result, a free 3' OH group is generated. This can be extended either enzymatically or chemically, to incorporate a detectable label. For example, enzymatic extension may occur. In a preferred embodiment, a non-templated extension occurs, for example, through the use of terminal transferase. Thus, for example, a modified dNTP may be incorporated, wherein the modification comprises the presence of a primary label such as a fluor, or a secondary label such as biotin, followed by the addition of a labeled streptavidin, for example.
Similarly, the addition of a template (e.g. a secondary target sequence that will hybridize to the detection sequence attached to the bead) allows the use of any number of reacfions as outlined herein, such as simple extension, SBE, pyrosequencing, OLA, etc. Again, this generally (but not always) utilizes the incorporation of a label into the growing strand.

Alternatively, as will be appreciated by those in the art, chemical labelling or extension methods may be used to label the 3' OH group.

As for all the combination methods, there are several advantages to this method. First of all, the absence of any label on the surface prior to cleavage allows a high signal-to-noise ratio. Additionally, the signalling probe need not contain any labels, thus making synthesis easier. Furthermore, because the target-specific por6on of the signalling probe is removed during the assay, the remaining detection sequence can be any sequence. This allows the use of a common sequence for all beads; even if different reactions are carried out in parallel on the array, the post-cleavage detection can be idenfical for all assays, thus requiring only one set of reagents. As will be appredated by those In the art, it Is also possible to have different detection sequences if required. In addition, since the label is attached post-cleavage, there is a great deal of flexibility In the type of label that may be incorporated. This can lead to significant signal ampiification; for example, the use of highly labeled streptavidin bound to a biofin on the detection sequence can give an Increased signal per detection sequence. Similarly, the use of enzyme labels such as alkaline phosphatase or horseradish peroxidase allow signal amplification as well.

A further advantage is the two-fold specifidty that is built into the assay.
By requiring specifidly at the cleavage step, followed by spedfidty at the extension step, increased signal-to-noise ratios are seen.
As will be appreciated by those in the art, whqe generally described as a solid phase assay, this readion may also be done in soludon; this Is similar to the solution-based SBE
readions, wherein the detection sequence serves as the extension primer. This assay also may be performed witlt an extension primer/adaptor oligonucleotide as described for solution-based SBE
assays. It should be noted that the arrays used to detect the invasive cleavage/extension reactions may be of any type, including, but not iimited to, spotted and printed arrays, photolithographic arrays, and bead arrays.
Combination of listafion and extension In a preferred embodiment, OLA and SBE are combined, as Is sometlmes referred to as "Genetic B#"
anaiysis and described in Nikforov et al., Nticleic Acid Res. 22:4167 (1994).
In this embodiment, the two ligation probes do not hybrkRze adjacwvtly;
rather, they are separated by one or more bases. The addition of dNTPs and a polymenme, In addition to the ligation probes and the ligase, results in an extended, ligated probe. As for SBE, the dNTPs may carry different labels, or separate reactions can be run, If the SBE
portlon of the reaction is used for genotyping. Alternatively, if the ligation portion of the reactlon Is used for genotyping, efther no extension occurs due to mismatch of the 3' base (such that the polymerase will not extend it), or no ligafion occurs due to mismatch of the 5' base. As will be appreciated by those In the art, the reacHon products are assayed using microsphere arrays. Again, as outlined herein, the assays may be solution based assays, with the ligated, extended probes being added to a microsphere array, or solid-phase assays. In additlon, the unextended, unligated primers may be removed priorto detection as needed, as is outlined herein. Furthermore, adapter sequences may also be used as outlined herein for OLA.

Combination of OLA and PCR
In a preferred embodiment, OLA and PCR are combined. As will be appreciated by those In the ert, the sequential order of the reaction Is variable. That is, in some embodiments It Is desired to perform the genotyping or OLA reac6on first followed by PCR amplification. In an aftemafve embodiment, It is desirable to first amplify the target i.e. by PCR followed by the OLA assay.

In a preferred embodiment, this technique is done on bead arrays.
Combination of competitive hybridization and ligation In a preferred embodiment, a combinafion of compefifive hybridizafion and ligation is done. As will be appreciated by those in the art, this technique is OLA as described above, with at least two sets of probes comprising different bases in the readout posifion. By running the reactions under conditions that favor hybridization complexes with perfect matches, different alleles may be distinguished.

In one embodiment, LCR is used to genotype a single genomic locus by incorporating two sets of two optically labeled AS oligonucleotides and a detecfion oligonucleofide in the ligation reaction. The oligonucleofide ligation step discriminates between the AS oligonucleofides through the efficiency of ligation between an oligonucleofide with a correct match with the target nucleic acid versus a mismatch base in the target nucleic acid at the ligation site. Accordingly, a detecfion oligonucleofide ligates efficiently to an AS oligonucleofide if there is complete base pairing at the ligafion site. One 3' oligonucleofide (T base at 5' end) is opfically labeled with FAM (green fluorescent dye) and the other 3' oligonucleofide (C base at 5' end) is labelled with TMR (yellow fluorescent dye). An A base in the target nucleic acid base pairs with the corresponding T resulfing in efficient ligation of the FAM-labeled oligonucleotide. A G base in the target nucleic acid results in ligation of the TMR-labeled oligonucleotide. TMR and FAM have disfinct emission spectrums. Accordingly, the wavelength of the oligonucleofide ligated to the 5' detection oligonucleotide indicates the nucleofide and thus the genotype of the target nucleic acid.

In a preferred embodiment, this technique is done on bead arrays.
Combinafion of competitive hybridizafion and invasive cleavage In a preferred embodiment, a combination of competitive hybridizafion and invasive cleavage is done.
As will be appreciated by those in the art, this technique is invasive cleavage as described above, with at least two sets of probes (either the invader probes or the signalling probes) comprising different bases in the readout posifion. By running the reacfions under conditions that favor hybridizafion complexes with perfect matches, different alleles may be disfinguished.

In a preferred embodiment, this technique is done on bead arrays.

In addition to the amplification and genotyping embodiments disclosed herein, the present invention further provides compositions and methods for nucleic acid sequencing.

SEQUENCING

The present invention is directed to the sequencing of nucleic acids, particularly DNA, by synthesizing nucleic acids using the target sequence (i.e. the nucleic acid for which the sequence is determined) as a template. These methods can be generally described as follows. A target sequence is attached to a solid support, either directly or indirectly, as outlined below. The target sequence comprises a first domain and an adjacent second domain comprising target positions for which sequence information is desired. A sequencing primer is hybridized to the first domain of the target sequence, and an extension enzyme is added, such as a polymerase or a ligase, as outlined below. After the addition of each base, the identity of each newly added base is determined prior to adding the next base. This can be done in a variety of ways, including controlling the reaction rate and using a fast detector, such that the newly added bases are identified in real time. Alternatively, the addition of nucleotides is controlled by reversible chain termination, for example through the use of photocleavable blocking groups. Alternatively, the addition of nucleotides is controlled, so that the reaction is limited to one or a few bases at a time. The reaction is restarted after each cycle of addition and reading. Alternatively, the addition of nucleotides is accomplished by carrying out a ligation reaction with oligonucleotides comprising chain terminating oligonucleotides. Preferred methods of sequencing-by-synthesis include, but are not limited to, pyrosequencing, reversible-chain termination sequencing, fime-resolved sequencing, ligation sequencing, and single-molecule analysis, all of which are described below.

The advantages of these "sequencing-by-synthesis" reactions can be augmented through the use of array techniques that allow very high density arrays to be made rapidly and inexpensively, thus allowing rapid and inexpensive nucleic acid sequencing. By "array techniques"
is meant techniques that allow for analysis of a plurality of nucleic acids in an array format.
The maximum number of nucleic acids is limited only by the number of discrete loci on a parficular array platform. As is more fully outlined below, a number of different array formats can be used.

The methods of the invention find particular use in sequencing a target nucleic acid sequence, i.e.
identifying the sequence of a target base or target bases in a target nucleic acid, which can ultimately be used to determine the sequence of long nucleic acids.

As is outlined herein, the target sequence comprises positions for which sequence information is desired, generally referred to herein as the "target positions". In one embodiment, a single target position is elucidated; in a preferred embodiment, a plurality of target positions are elucidated. In general, the plurality of nucleotides in the target positions are contiguous with each other, although in some circumstances they may be separated by one or more nucleotides. By "plurality" as used herein is meant at least two. As used herein, the base which basepairs with the target position base in a hybrid is termed the "sequence position". That is, as more fully outlined below, the extension of a sequence primer results in nucleotides being added in the sequence positions, that are perfectly complementary to the nucleotides in the target positions. As will be appreciated by one of ordinary skill in the art, identification of a plurality of target positions in a target nucleotide sequence results in the determination of the nucleotide sequence of the target nucleotide sequence.

As will be appreciated by one of ordinary skill in the art, this system can take on a number of different configurafions, depending on the sequencing method used, the method of attaching a target sequence to a surface, etc. In general, the methods of the invention rely on the attachment of different target sequences to a solid support (which, as outlined below, can be accomplished in a variety of ways) to form an array. The target sequences comprise at least two domains: a first domain, for which sequence information is not desired, and to which a sequencing primer can hybridize, and a second domain, adjacent to the first domain, comprising the target positions for sequencing. A sequencing primer is hybridized to the target sequence, forming a hybridization complex, and then the sequencing primer is enzymatically extended by the addition of a first nucleofide into the first sequence position of the primer. This first nucleotide is then iden6fied, as is outlined below, and then the process is repeated, to add nucleotides to the second, third, fourth, etc. sequence positions. The exact methods depend on the sequencing technique utilized, as outlined below.

Once the target sequence is associated onto the array as outlined below, the target sequence can be used in a variety of sequencing by synthesis reactions. These reactions are generally classified into several categories, outlined below.

SEQUENCING BY SYNTHESIS
As outlined herein, a number of sequencing by synthesis reactions are used to elucidate the identity of a plurality of bases at target positions within the target sequence. All of these reactions rely on the use of a target sequence comprising at least two domains; a first domain to which a sequencing primer will hybridize, and an adjacent second domain, for which sequence information is desired. Upon formation of the assay complex, extension enzymes are used to add dNTPs to the sequencing primer, and each addition of dNTP is "read" to determine the identity of the added dNTP. This may proceed for many cycles.

Pyroseguencing In a preferred embodiment, pyrosequencing methods are done to sequence the nucleic acids. As outlined above, pyrosequencing is an extension method that can be used to add one or more nucleotides to the target positions. Pyrosequencing relies on the detection of a reaction product, pyrophosphate (PPi), produced during the addifion of an NTP to a growing oligonucleo6de chain, rather than on a label attached to the nucleotide. One molecule of PPi is produced per dNTP added to the extension primer. The detection of the PPi produced during the reac6on is monitored using secondary enzymes; for example, preferred embodiments utilize secondary enzymes that convert the PPi into ATP, which also may be detected in a variety of ways, for example through a chemiluminescent reaction using luciferase and luciferin, or by the detection of NADPH. Thus, by running sequenfial reactions with each of the nucleotides, and monitoring the reaction products, the identity of the added base is determined.

Accordingly, the present invention provides methods of pyrosequencing on arrays; the arrays may be any number of different array configurations and substrates, as outlined herein, with microsphere arrays being paracularly preferred. In this embodiment, the target sequence comprises a first domain that is substantially complementary to a sequencing primer, and an adjacent second domain that comprises a plurality of target positions. By "sequencing primer" herein is meant a nucleic acid that is substantially complementary to the first target domain, with perfect complementarity being preferred.
As will be appreciated by those in the art, the length of the sequencing primer will vary with the condifions used. In general, the sequencing primer ranges from about 6 to about 500 or more basepairs in length, with from about 8 to about 100 being preferred, and from about 10 to about 25 being especially preferred.

Once the sequencing primer is added and hybridized to the target sequence to form a first hybridization complex (also somefimes referred to herein as an "assay complex"), the system is ready to inifiate sequencing-by-synthesis. The methods described below make reference to the use of fiber optic bundle substrates with associated microspheres, but as will be appreciated by those in the art, any number of other substrates or solid supports may be used, or arrays that do not comprise microspheres.

The reaction is initiated by introducing the substrate comprising the hybridization complex comprising the target sequence (i.e. the array) to a solution comprising a first nucleotide, generally comprising deoxynucleoside-triphosphates (dNTPs). Generally, the dNTPs comprise dATP, dTTP, dCTP and dGTP. The nucleotides may be naturally occurring, such as deoxynucleotides, or non-naturally occurring, such as chain terminating nucleofides including dideoxynucleofides, as long as the enzymes used in the sequencing/detection reactions are still capable of recognizing the analogs. In addition, as more fully outlined below, for example in other sequencing-by-synthesis reactions, the nucleotides may comprise labels. The different dNTPs are added either to separate aliquots of the hybridization complex or preferably sequentially to the hybridization complex, as is more fully outlined below. In some embodiments it is important that the hybridization complex be exposed to a single type of dNTP
at a fime.

In addition, as will be appreciated by those in the art, the extension reactions of the present invenfion allow the precise incorporafion of modified bases into a growing nucleic acid strand. Thus, any number of modified nucleofides may be incorporated for any number of reasons, including probing structure-function relationships (e.g. DNA:DNA or DNA:protein interacfions), cleaving the nucleic acid, crosslinking the nucleic acid, incorporate mismatches, etc.

In addition to a first nucleotide, the solution also comprises an extension enzyme, generally a DNA

polymerase. Suitable DNA polymerases include, but are not iimited to, the Kienow fragment of DNA
polymerase I, SEQUENASE 1.0 and SEQUENASE 2.0 (U.S. Biochemical), T5 DNA
polymerase and Pht29 DNA polymerase. If the dNTP is complementary to the base of the target sequence adjacent to the extension primer, the extension enzyme will add it to the extension primer, releasing pyrophosphate (PPi). Thus, the extension primer is modified, i.e. extended, to form a modified primer, sometimes referred to herein as a'newly synthesized strand'. The incorporaflon of a dNTP into a newly synthesized nucleic acid strand releases PPi, one molecule of PPI per dNTP incorporated.

The release of pyrophosphate (PPQ during the DNA poiymerase reacfion can be quantitatively measured by many different methods and a number of enzymatic methods have been described; see Reeves et al., Anal. Biochem. 28282 (1969); Guiliory et al., Anal. Biochem.
39:170 (1971); Johnson et al., Anal. Biochem.15:273 (1968); Cook et al., Anal. Biochem. 91:557 (1978);
Drake et al., Anal.
Biochem. 94:117 (1979); Ronaghl et al., Science 281:363 (1998); Barshop et al., Anal. Biochem.
197(1):266-272 (1991) W093/23564; WO 98/28440; W098/13523; Nyren et al., Anal.
Biochem.
151:504 (1985). The latter method allows continuous monitoring of PPi and has been termed ELIDA (Enzymatlc Luminometric Inorganic Pyrophosphate Detection Assay). In a preferred embodiment, the PPi is detected u6lizing UDP-glucose pyrophosphorylase, phosphoglucomutase and glucose 6-phosphate dehydrogenase.
See Justesen, et al., Anal. Biochem. 207(1):90-93 (1992); Lust et al., Clin. Chem. Acta 66(2)241 (1976); and Johnson et al., Anal. Biochem. 26:137 (1968). This reactlon produces NADPH which can be detected fluoremetrically.
A preferred embodiment utiiizes any method which can result in the generation of an opticai signal, with preferred embodiments utilizing the generation of a chemiluminescent or fluorescent signal.
Generally, these methods rely on secondary enzymes to detect the PPi; these methods generally rely on enzymes that will convert PPi Into ATP, which can then be detected. A
preferred method monitors the creation of PPi by the conversion of PPI to ATP by the enzyme sulfurylase, and the subsequent production of visible light by firefly iuciferase (see Ronaghi et al., supra, and Ban3hop, supra). In this method, the four deoxynucleotides (dATP, dGTP, dCTP and dTTP; collectively dNTPs) are added stepwise to a partial duplex comprising a sequendng primer hybridized to a single stranded DNA
template and incubated with DNA polymerase, ATP sulfurylase (and its substrate, adenosine 5'-3 0 phosphosulphate (APS)) luciferase (and Its substrate luciferin), and optlonaiy a nucleotide-degrading enzyme such as apyrase. A dNTP is only incorporated into the growing DNA
strand if it is complementary to the base in the template strand. The synthesis of DNA is accompanied by the release of PPi equal in molarity to the incorporated dNTP. The PPi is converted to ATP and the light generated by the luciferase Is direcUy propor6onai to the amount of ATP. In some cases the unincorporated dNTPs and the produced ATP are degraded between each cycle by the nucleoflde degrading enzyme.

As will be appreciated by those in the art, if the target sequence comprises two or more of the same nucieotide in a row, more than one dNTP will be incorporated; however, the amount of PPI generated Is directly proportionai to the number of dNTPs incorporated and thus these sequences can bedetected.

In addition, in a preferred embodiment, the dATP that is added to the reaction m"ature is an analog that can be incorporated by the DNA polymerase into the growing oligonucieotide strand, but will not serve as a substrate for the second enzyme; for example, certain thiol-containing dATP analogs find particuiaruse.

Accordingly, a preferred embodiment of the methods of the invention is as follows. A substrate comprising microspheres containing the target sequences and extension primers, forming hybridization complexes, is dipped or contacted with a volume (reaction chamber or well) comprising a single type of dNTP, an extension enzyme, and the reagents and enzymes necessary to detect PPI. If the dNTP is complementary to the base of the target portion of the target sequence adjacent to the extension primer, the dNTP is added, releasing PPi and generating detectable light, which Is detected as generally described in U.S.S.N.s 09/151,877 and 09/189,543, and PCT
U598/09163.
If the dNTP is not complementary, no detectable signal resuits.
The substrate is then contacted with a second reaction chamber comprising a different dNTP and the additional components of the assay. This process is repeated to generate a readout of the sequence of the target sequence.

In a preferred embodiment, washing steps, i.e. the use of washing chambers, may be done in between the dNTP react+on chambers, as required. These washing chambers may opfionaiiy comprise a nucieotide-degrading enzyme, to remove any unreacted dNTP and decreasing the background signal, as is described in WO 98/28440. In a preferred embodiment a flow cell is used as a reaction chamber; following each reaction the unreacted dNTP
is washed away and may be replaced with an additionai dNTP to be examined.

As will be appreciated by those in the art, the system can be configured in a variety of ways, including both a linear progression or a circular one; for example, four substrates may be used that each can dip Into one of four reaction chambers arrayed in a circular pattern. Each cycle of sequencing and reading Is followed by a 90 degree rotation, so that each substrate then dips into the next reaction well. This allows a condnuous series of sequencing reactions on multipie substrates in parallel.

In a preferred embodiment, one or more internal control sequences are used.
That is, at least one microsphere in the array comprises a known sequence that can be used to verify that the reactions are proceeding correctiy. In a preferred embodiment, at least four control sequences are used, each of which has a different nucleotide at each position: the first control sequence will have an adenosine at position 1, the second wiil have a cytosine, the thircl a guanosine, and the fourth a thymidine, thus ensuring that at least one control sequence Is "lighting up" at each step to serve as an intemal control.
In a preferred embodiment, the reaction Is run for a number of cycies un6i the signahto-noise ratio becomes low, generally from 20 to 70 cycles or more, with from about 30 to 50 being standard. In some embodiments, this is suffident for the purposes of the experiment; for example, for the detectlon of certain mutations, including single nucleotide polymorphisms (SNPs), the experiment Is designed such that the inffial round of sequencing gives the desired Information. In other embodiments, it Is desirable to sequence longer targets, for example in excess of hundreds of bases. In this appiication, additional rounds of sequencing can be done.

For example, after a certain number of cydes, ii: Is possible to stop the reaction, remove the newly synthesized strand using either a thermal step or a chemical wash, and start the reac6on over, using for example the sequence information that was previously generated to make a new extension primer that will hybridize to the first target portlon of the target sequence. That Is, the sequence information generated in the first round is transferred to an oligonucieotide synthesizer, and a second extension primer is made for a second round of sequencing. In this way, muitipie overlapping rounds of sequencing are used to generate long sequences from template nucieic aoid strands. Aiternativeiy, when a single target sequence contains a number of mutationai "hot spots', primers can be generated using the known sequences In between these hot spots.

Additionaliy, the methods of the invention find use in the decoding of random microsphere arrays.
That is, as descxibed in U.S.S.N. 09/189,543, nucleic acids can be used as bead identiflers. By using sequencing-by-synthesis to read out the sequence of the nucleic acids, the beads can be decoded in a highly parallel fashion.

In addition, the methods find use in simuitaneous analysis of muitipie target sequence positions on a single array. For example, four separate sequence analysis reactions are performed. In the first reaction, posiBons containing a particuiar nucieotlde ('A', for example) in the target sequence are analyzed. In three other reactions, C, G, and T are analyzed. An advantage of anaiyzing one base per reaction is that the baseline or background Is flattened for the three bases exciuded from the reaction. Therefore, the signal is more easily detected and the sensitivity of the assay Is Increased.
Aitemativeiy, each of the four sequencing reactlons (A, G, C and T) can be performed simuitaneousiy with a nested set of primers providing a significant advantage in that primer synthesis can be made more efficient.

In another preferred embodiment each probe is represented by muitipie beads in the array (see U.S.S.N. 09/287,573, fiied April 6, 1999). As a resuit, each experiment can be replicated many times in parallel. As outlined below, averaging the signal from each respective probe in an experiment also allows for improved signal to noise and increases the sensitivity of detecting subtle perturbations in signal intensity patterns. The use of redundancy and comparing the patterns obtained from two different samples (e.g. a reference and an unknown), results in highly paralleled and comparative sequence analysis that can be performed on complex nucleic acid samples.

As outlined herein, the pyrosequencing systems may be configured in a variety of ways; for example, the target sequence may be attached to the array (e.g. the beads) in a variety of ways, including the direct attachment of the target sequence to the array; the use of a capture probe with a separate extension probe; the use of a capture extender probe, a capture probe and a separate extension probe; the use of adapter sequences in the target sequence with capture and extension probes; and the use of a capture probe that also serves as the extension probe.

In addition, as will be appreciated by those in the art, the target sequence may comprise any number of sets of different first and second target domains; that is, depending on the number of target positions that may be elucidated at a time, there may be several "rounds" of sequencing occuring, each time using a different target domain.

One additional benefit of pyrosequencing for genotyping purposes is that since the reaction does not rely on the incorporation of labels into a growing chain, the unreacted extension primers need not be removed.

Thus, pyrosequencing kits and reactions require, in no particularly order, arrays comprising capture probes, sequencing primers, an extension enzyme, and secondary enzymes and reactants for the detection of PPi, generally comprising enzymes to convert PPi into ATP (or other NTPs), and enzymes and reactants to detect ATP.

Attachment of enzymes to arrays In a preferred embodiment, particularly when secondary enzymes (i.e. enzymes other than extension enzymes) are used in the reaction, the enzyme(s) may be attached, preferably through the use of flexible linkers, to the sites on the array, e.g. the beads. For example, when pyrosequencing is done, one embodiment utilizes detection based on the generation of a chemiluminescent signal in the "zone"
around the bead. By attaching the secondary enzymes required to generate the signal, an increased concentration of the required enzymes is obtained in the immediate vicinity of the reaction, thus allowing for the use of less enzyme and faster reaction rates for detection.
Thus, preferred embodiments utilize the attachment, preferably covalently (although as will be appreciated by those in the art, other attachment mechanisms may be used), of the non-extension secondary enzymes used to generate the signal. In some embodiments, the extension enzyme (e.g. the polymerase) may be attached as well, although this is not generally preferred.

The attachment of enzymes to array sites, particularly beads, Is outGned in U.S.S.N. 09/287,573, hereby Incorporated by reference, and will be appreciated by those in the art.
In general, the use of flexible linkers are preferred, as this allows the enzymes to interacYwlth the substrates. However, for some types of attachment, linkers are not needed. Attachment proceeds on the basis of the composition of the array site (i.e. either the substrate or the bead, depending on which array system is used) and the compostiion of the enzyme. In a preferred embodiment, depending on the composition of the array site (e.g. the bead), it will contain chemical functional groups for subsequent attachment of other moieties. For example, beads comprising a variety of chemical funcdonal groups such as amines are commercially available. Preferred functional groups for attachment are amino groups, carboxy groups, oxo groups and thiol groups, with amino groups being partlcularly preferred. Using these functional groups, the enzymes can be attached using functional groups on the enzymes. For example, enzymes containing amino groups can be attached to partlcles comprising amino groups, for example using linkers as are known In the art for example, homo-or hetero-bifunctional linkers as are well known (see 1994 Pierce Chemical Company catalog, technical section on cross-linkers, pages 155-200.

Reversible Chain Termination Methods in a preferred embodiment, the sequencing-by-synthesis method utiiized is reversible chain termination. In this embodiment, the rate of addi6on of dNTPs is controlled by using nucieotlde analogs that contain a removable protecting group at the 3' position of the dNTP. The presence of the protecting group prevents further addition of dNTPs at the 3' end, thus aliowing tlme for detection of the nucieotide added (for example, utilizing a labeled dNTP). After acquisition of the identity of the dNTP added, the protecting group is removed and the cycle repeated. In this way, dNTPs are added one at a time to the sequencing primer to allow eiucidation of the nucleotides at the target positlons.
See U.S. Patent Nos. 5,902,723; 5,547,839; Metzker et al., Nucl. Add Res.
22(20):4259 (1994);
Canard et ai., Gene 148(1):1-6 (1994); Dyatkina et al., Nucleic Acid Symp.
Ser. 18:117-120 (1987).
Accordingly, the present invention provides methods and compositions for reversible chain termination sequencing-by-synthesis. Similar to pyrosequencing, the reaction requires the hybridization of a substantially complementary sequencing primer to a first target domain of a target sequence to form an assay complex.

The reaction is ini6ated by introducing the assay complex comprising the target sequence (i.e. the array) to a soiution comprising a first nucieotide analog. By "nucleotide analog' in this context herein is meant a deoxynucleoside-triphosphate (also called deoxynucieotides or dNTPs, i.e. dATP, dTTP, dCTP and dGTP), that is further derivatized to be reversibly chain terminating. As wili be appreciated by those in the art, any number of nucleotide analogs may be used, as long as a poiymerase enzyme wiil stiii incorporate the nucleotide at the sequence posi4on. Preferred embodiments utilize 3'-O-methyl-dNTPs (with photolytic removal of the protecting group), 3'-substituted-2'-dNTPs that contain anthranylic derivatives that are fluorescent (with alkali or enzymatic treatment for removal of the protecting group). The latter has the advantage that the protecting group is also the fluorescent label;
upon cleavage, the label is also removed, which may serve to generally lower the background of the assay as well.

Again, the system may be configured and/or ufilized in a number of ways. In a preferred embodiment, a set of nucleotide analogs such as derivatized dATP, derivatized dCTP, derivatized dGTP and derivatized dTTP is used, each with a different detectable and resolvable label, as outlined below.
Thus, the identification of the base at the first sequencing position can be ascertained by the presence of the unique label.

Alternatively, a single label is used but the reactions are done sequentially.
That is, the substrate comprising the array is first contacted with a reaction mixture of an extension enzyme and a single type of base with a first label, for example ddATP. The incorporation of the ddATP is monitored at each site on the array. The substrate is then contacted (with opfional washing steps as needed) to a second reaction mixture comprising the extension enzyme and a second nucleofide, for example ddTTP. The reaction is then monitored; this can be repeated for each target position.

Once each reaction has been completed and the identification of the base at the sequencing position is ascertained, the terminafing protecting group is removed, e.g. cleaved, leaving a free 3' end to repeat the sequence, using an extension enzyme to add a base to the 3' end of the sequencing primer when it is hybridized to the target sequence. As will be appreciated by those in the art, the cleavage conditions will vary with the protecting group chosen.

In a preferred embodiment, the nucleotide analogs comprise a detectable label as described herein, and this may be a primary label (directly detectable) or a secondary label (indirectly detectable).

In addition to a first nucleotide, the solution also comprises an extension enzyme, generally a DNA
polymerase, as outlined above for pyrosequencing.

In a preferred embodiment, the protecting group also comprises a label. That is, as outlined in Canard et al., supra, the protecting group can serve as either a primary or secondary label, with the former being preferred. This is particularly preferred as the removal of the label at each round results in less background noise, less quenching and less crosstalk.

In this way, reversible chain termination sequencing is accomplished.
Time-resolved sequencing In a preferred embodiment, time-resoived sequencing is done. This embodiment relies on controlling the reacaon rate of the extension reaction and/or using a fast imaging system.
Basically, the method involves a simple extension reaction that Is either'siowed down', or imaged using a fast system, or both. What is important Is that the rate of poiymeriza6on (extension) is significantiy slower than the rate of image capture.

To allow for real time sequencing, parameters such as the speed of the detector (millisecond speed Is preferred), and rate of poiymerization wili be controlled such that the rate of poiymerization Is significantiy slower than the rate of image capture. Polymerization rates on the order of kiiobases per minute (e.g. -10 milliseconds/nucieotide), which can be adjusted, should allow a sufficientiy wide window to find conditions where the sequential addi6on of two nucieotides can be resolved. The DNA
poiymerization reaction, which has been studied intensively, can easily be reconstituted In vitro and controlled by varying a number of parameters inciuding reacction temperature and the concentration of nucieotide triphosphates.

In addition, the polymerase can be applied to the primer-tempiate complex prior to initiating the reaction. This serves to synchronize the reaction. Numerous polymerases are available. Some examples include, but are not iimited to polymerases with 3' to 5' exonuciease activity, other nuciease activities, polymerases with different processMty, affinities for modified and unmodified nucleotide triphosphates, temperature optima, stability, and the like.

Thus, In this embodiment, the reaction proceeds as outlined above. The target sequence, comprising a first domain that will hybridize to a sequencing primer and a second domain comprising a plurality of target positions, is attached to an array as outlined below. The sequencing primers are added, along with an extension enzyme, as outiined herein, and dNTPs are added. Again, as outlined above, either four differently labeled dNTPs may be used simuitaneousiy or, four different sequential reactions with a single label are done. In general, the dNTPs comprise either a primary or a secondary label, as outlined above.

In a preferred embodiment, the extension enzyme is one that is relatively "siow". This may be accomplished in several ways. In one embodiment, polymerase variants are used that have a lower poiymerization rate than wiid-type enzymes. Altematively, the reaction rate may be controlled by varying the temperature and the concentration of dNTPs.

In a preferred embodiment, a fast (millisecond) high-sensitlvity Imaging system is used.

In one embodiment, DNA poiymerizatlon (extension) is monitored using light scattering, as is outiined in Johnson et al., Anal. Biochem. 136(1):192 (1984).

ATTACHMENT OF TARGET SEQUENCES TO ARRAYS
As is generally described herein, there are a variety of methods that can be used to attach target sequences to the solid supports of the invention, particularly to the microspheres that are distributed on a surface of a substrate. Most of these methods generally rely on capture probes attached to the array. However, the attachment may be direct or indirect. Direct attachment includes those situations wherein an endogeneous portion of the target sequence hybridizes to the capture probe, or where the target sequence has been manipulated to contain exogeneous adapter sequences that are added to the target sequence, for example during an amplification reacfion. Indirect attachment utilizes one or more secondary probes, termed a "capture extender probe" as outlined herein.

In a preferred embodiment, direct attachment is done, as is generally depicted in Figure IA. In this embodiment, the target sequence comprises a first target domain that hybridizes to all or part of the capture probe.

In a preferred embodiment, direct attachment is accomplished through the use of adapters. The adapter is a chemical moiety that allows one to address the products of a reaction to a solid surface.
The type of reaction includes the amplificafion, genotyping and sequencing reacfions disclosed herein.
The adapter chemical moiety is independent of the reaction. Because the adapters are independent of the reaction, sets of adapters can be reused to create a "universal" array that can detect a variety of products from a reaction by attaching the set of adapters that address to specific locations within the array to different reactants.

Typically, the adapter and the capture probe on an array are binding partners, as defined herein.
Although the use of other binding partners are possible, preferred embodiments utilize nucleic acid adapters that are non-complementary to any reactants or target sequences, but are substantially complementary to all or part of the capture probe on the array.

Thus, an "adapter sequence" is a nucleic acid that is generally not native to the target sequence, i.e. is exogeneous, but is added or attached to the target sequence. It should be noted that in this context, the "target sequence" can include the primary sample target sequence, or can be a derivative target such as a reactant or product of the reactions outlined herein; thus for example, the target sequence can be a PCR product, a first ligafion probe or a ligated probe in an OLA
reacfion, etc.

As will be appreciated by those in the art, the attachment, or joining, of the adapter sequence to the target sequence can be done in a variety of ways. In a preferred embodiment, the adapter sequences are added to the primers of the reaction (extension primers, amplification primers, readout probes, sequencing primers, Rolling Circle primers, etc.) during the chemical synthesis of the primers. The adapter then gets added to the reaction product during the reacfion; for example, the primer gets extended using a polymerase to form the new target sequence that now contains an adapter sequence. Altematively, the adapter sequences can be added enzymatically.
Furthermore, the adapter can be attached to the target after synthesis; this post-synthesis attachment could be either covalent or non-covalent.

In this embodiment, one or more of the amplification primers comprises a first portion compnsing the adapter sequence and a second portion comprising the primer sequence.
Extending the ampiification primer as is well known in the art results in target sequences that comprise the adapter sequences.
The adapter sequences are designed to be substan6ally complementary to capture probes.

In addition, as wiil be appreciated by those in the art, the adapter can be attached either on the 3' or 5' ends, or in an intemal position. For example, the adapter may be the detection sequence of an invasive cleavage probe. In the case of Rolling Circle probes, the adapter can be contained within the section between the probe ends. Adapters can also be attached to aptamers.
Aptamers are nudeic acids that can be made to bind to virtually any target analyte; see Bock et al., Nature 355:564 (1992);
Femulok et al., Current Op. Chem. Biol. 2:230 (1998); and U.S. Patents 5,270,163, 5,475,096, 5,567,588, 5,595,877, 5,637,459, 5,683,867,5,705,337, and related patents.
In additlon, as oudined below, the adapter can be attached to non-nucleic acid target analytes as well.

In one embodiment, a set of probes is hybridized to a target sequence; each probe is complementary to a different region of a single target but each contains the same adapter.
Using a poly-T bead, the mRNA target is pulled out of the sampie with the probes attached.
Dehybridizing the probes attached to the target sequence and rehybridizing them to an array containing the capture probes complementary to the adapter sequences results in binding to the array. All adapters that have boudn to the same target mRNA will bind to the same location on the array.

In a preferred embodiment, indirect attachment of the target sequence to the array is done, through the use of capture extender probes. "Capture extender" probes are generally depicted in Figure IC, and other figures, and have a first portion that wili hybrid'¾e to all or part of the capture probe, and a second portion that will hybridize to a first pordon of the target sequence.
Two capture extender probes may also be used. This has generally been done to stabilize assay complexes for example when the target sequence is large, or when large amplifier probes (partlcularly branched or dendrimer amplifier probes) are used.

When only capture probes are utiiized, it is necessary to have unique capture probes for each target sequence; that is, the surface must be customized to contain unique capture probes; e.g. each bead comprises a different capture probe. In general, only a single type of capture probe should be bound to a bead; however, different beads should contain different capture probes so that different target sequences bind to different beads.

Alternatively, the use of adapter sequences and capture extender probes allow the creation of more "universal" surfaces. In a preferred embodiment, an array of different and usually artificial capture probes are made; that is, the capture probes do not have complementarity to known target sequences.
The adapter sequences can then be added to any target sequences, or soluble capture extender probes are made; this allows the manufacture of only one kind of array, with the user able to customize the array through the use of adapter sequences or capture extender probes. This then allows the generation of customized soluble probes, which as will be appreciated by those in the art is generally simpler and less costly.

As will be appreciated by those in the art, the length of the adapter sequences will vary, depending on the desired "strength" of binding and the number of different adapters desired. In a preferred embodiment, adapter sequences range from about 6 to about 500 basepairs in length, with from about 8 to about 100 being preferred, and from about 10 to about 25 being particularly preferred.

In one embodiment, microsphere arrays containing a single type of capture probe are made; in this embodiment, the capture extender probes are added to the beads prior to loading on the array. The capture extender probes may be addifionally fixed or crosslinked, as necessary.

In a preferred embodiment, as outlined in Figure 1 B, the capture probe comprises the sequencing primer; that is, after hybridization to the target sequence, it is the capture probe itself that is extended during the synthesis reaction.

In one embodiment, capture probes are not used, and the target sequences are attached directly to the sites on the array. For example, libraries of clonal nucleic acids, including DNA and RNA, are used. In this embodiment, individual nucleic acids are prepared, generally using conventional methods (including, but not limited to, propagation in plasmid or phage vectors, amplificafion techniques including PCR, etc.). The nucleic acids are preferably arrayed in some format, such as a microtiter plate format, and either spotted or beads are added for attachment of the libraries.

Attachment of the clonal libraries (or any of the nucleic acids outlined herein) may be done in a variety of ways, as will be appreciated by those in the art, including, but not limited to, chemical or affinity capture (for example, including the incorporafion of derivatized nucleotides such as AminoLink or biotinylated nucleofides that can then be used to attach the nucleic acid to a surface, as well as affinity capture by hybridization), cross-linking, and electrostatic attachment, etc.

In a preferred embodiment, affinity capture is used to attach the clonal nucleic acids to the surface.
For example, cloned nucleic acids can be derivatized, for example with one member of a binding pair, and the beads derivatized with the other member of a binding pair. Suitable binding pairs are as described herein for secondary labels and IBUDBL pairs. For example, the cloned nucleic acids may be biotinylated (for example using enzymafic incorporate of biofinylated nucleofides, for by photoactivated cross-lin(ing of biofin). Biotinylated nucleic acids can then be captured on streptavidin-coated beads, as is known in the art. Similarly, other hapten-receptor combinations can be used, such as digoxigenin and anti-digoxigenin antibodies. Alternafively, chemical groups can be added in the form of derivatized nucleotides, that can them be used to add the nucleic acid to the surface.
Preferred attachments are covalent, although even relatively weak interactions (i.e. non-covalent) can be sufficient to attach a nucleic acid to a surface, if there are muitiple sites of attachment per each nucleic acid. Thus, for example, electrostatic interactions can be used for attachment, for example by having beads carrying the opposite charge to the bioactive agent.

Similarly, affinity capture ufilizing hybridization can be used to attach cloned nucleic acids to beads.
For example, as is known in the art, polyA+RNA is roufinely captured by hybridization to oligo-dT
beads; this may include oligo-dT capture followed by a cross-linking step, such as psoralen crosslinking). If the nucleic acids of interest do not contain a polyA tract, one can be attached by polymerization with terminal transferase, or via ligation of an oligoA linker, as is known in the art.

Alternatively, chemical crosslinking may be done, for example by photoactivated crosslinking of thymidine to reactive groups, as is known in the art.

In general, special methods are required to decode clonal arrays, as is more fully outlined below.
ASSAY AND ARRAYS
All of the above compositions and methods are directed to the detection and/or quantification of the products of nucleic acid reactions. The detecfion systems of the present invention are based on the incorporation (or in some cases, of the deletion) of a detectable label into an assay complex on an array.

Accordingly, the present invention provides methods and compositions useful in the detection of nucleic acids. As will be appreciated by those in the art, the compositions of the invention can take on a wide variety of configurations, as is generally outlined in the Figures. As is more fully outlined below, preferred systems of the invention work as follows. A target nucleic acid sequence is attached (via hybridization) to an array site. This attachment can be either directly to a capture probe on the surface, through the use of adapters, or indirectly, using capture extender probes as outlined herein.
In some embodiments, the target sequence itself comprises the labels.
Alternatively, a label probe is then added, forming an assay complex. The attachment of the label probe may be direct (i.e.
hybridization to a por6on of the target sequence), or indirect (i.e.
hybridization to an amplifier probe that hybridizes to the target sequence), with all the required nucleic acids forming an assay complex.

Accordingly, the present invention provides array compositions comprising at least a first substrate with a surface comprising individuai sites. By "amay" or "biochip" herein is meant a plurality of nucieic acids In an array format; the size of the array wiii depend on the compositlon and end use of the array.
Nucleic acids arrays are known in the art, and can be classified in a number of ways; both ordered arrays (e.g. the ability to resolve chemistrles at discrete sites), and random arrays are included.
Ordered arrays include, but are not limited to, those made using photoi'dhography techniques (Affymetrix GeneChipTM), spottlng techniques (Synteni and others), printing techniques (Hewlett Packard and Rosetta), three dimensional "gei pad" arrays, etc. A preferred embodiment utliizes microspheres on a variety of substrates including fiber optic bundles, as are outlined In PCTs US98/21193, PCT US99/14387 and PCT US98/05025; W098/50782; and U.S.S.N.s 09287,573, 09/151,877, 09/256,943, 09/316,154, 60/119,323, 09/315,584.
While much of the discussion below is directed to the use of microsphere arrays on fiber optic bundles, any array format of nucieic acids on soUd supports may be utiiized.

Arrays containing from about 2 different bioactive agents (e.g. different beads, when beads are used) to many miiiions can be made, with very large arrays being possible.
Generally, the array wiil comprise from two to as many as a billion or more, depending on the size of the beads and the substrate, as well as the end use of the array, thus very high densily, high density, moderate density, low density and very low density arrays may be made. Preferred ranges for very high density arrays are from about 10,000,000 to about 2,000,000,000, with from about 100,000,000 to about 1,000,000,000 being preferred (all numbers being in square cm). High density arrays range about 100,000 to about 10,000,000, with from about 1,000,000 to about 5,000,000 being partlcuiariy preferred. Moderate density arrays range from about 10,000 to about 100,000 being parficuiariy preferred, and from about 20,000 to about 50,000 being especially preferred.
Low density arrays are generally less than 10,000, with from about 1,000 to about 5,000 being preferred. Very low denslty arrays are less than 1,000, with from about 10 to about 1000 being preferred, and from about 100 to about 500 being particuiarly preferred. In some embodiments, the compositions of the invention may not be in array format; that is, for some embodiments, compositions comprising a single bioactive agent may be made as well. In addition, in some arrays, muifipie substrates may be used, aither of different or identicai compositions. Thus for example, large arrays may comprise a piurality of smaller substrates.

In addition, one advantage of the present compositions is that particuiariy through the use of fiber optic technology, extremely high density arrays can be made. Thus for example, because beads of 200 pm or less (with beads of 200 nm possible) can be used, and very small fibers are known, it Is possible to have as many as 40,000 or more (in some instances,l million) different elements (e.g. fibers and beads) in a 1 mm' fiber optic bundle, with densities of greater than 25,000,000 individuai beads and fibers (again, In some instances as many as 50-100 miiiion) per 0.5 cm2 obtainable (4 mliion per square cm for 5 N center-to-center and 100 million per square cm for 1 p center-to-center).

By "substrate" or "soiid support" or other grammatical equivalents herein is meant any materiai that can be modified to contain discrete lndMdual sRes appropriate for the attachment or association of beads and is amenable to at least one detection method. As will be appreciated by those in the art, the number of possible substrates is very large. Possible substrates include, but are not limibed to, glass and modified or functionalized glass, plastics (including acrylics, polystyrene and copolymers of styrene and other materials, polypropylene, polyethyiene; polybutyiene, polyurethanes, Teflon, etc.), polysaccharides, nylon or nitrocelluiose, resins, silica or silica-based materials including silicon and modified silicon, carbon, metals, inorganic glasses, piastics, optical flber bundles, and a variety of other polymers. In general, the substrates allow optical detecdion and do not themselves appreciably fluoresce.

Generally the substrate is flat (planar), although as will be appreciated by those in the art; other configurations of substrates may be used as well; for example, three dimensional configurations can be used, for example by embedding the beads In a porous block of plastic that allows sample access to the beads and using a confocal microscope for detection. Similarly, the beads may be placed on the inside surface of a tube, for flow-through sample analysis to minimize sample volume. Preferred substrates Include optical fiber bundles as discussed below, and flat planar substrates such as glass, polystyrene and other plastic.s and acrylios.

In a preferred embodiment, the substrate is an optical fiber bundle or array, as is generally described in PCT US98/05025, and PCT US98/09163.
Preferred embodiments utilize preformed unitary fiber optic arrays. By "preformed unitary fiber optic array" herein Is meant an array of discrete individual fiber optic strands that are co-axially disposed and joined along their lengths. The fiber strands are generally Individually clad. However, one thing that distinguished a preformed unitary array from other fiber optic formats is that the fibers are not individuaiiy physically manipulatable; that Is, one strand generally cannot be physically separated at any point along its length from another fiber strand.
Generally, the array of array compositions of the invention can be configured In severai ways.
In a preferred embodiment, as Is more fully outlined below, a"one component" system Is used.
That is, a first substrate comprising a plurality of assay locations (sometimes also referred to herein as'assay wells"), such as a microtiter plate, is configured such that each assay location contains an individual array. That is, the assay location and the array locatlon are the same. For example, the plastic material of the microtiter plate can be formed to contain a piurality of `bead welis' in the bottom of each of the assay wells. Beads containing the capture probes of the invention can then be loaded Into the bead wells in each assay iocation as Is more fully described below.

Altemativeiy, a'two component" system can be used. In this embodiment, the individuai arnays are formed on a second substrate, which then can be fitted or "dipped" into the first microtiter plate substrate. A preferred embodiment utilizes fiber optic bundles as the individual arrays, generally with "bead wells" etched into one surface of each indMdual fiber, such that the beads containing the capture probes are loaded onto the end of the fiber optic bundle. The composite array thus comprises a number of individual arrays that are configured to fit within the wells of a microtiter plate.
By "composite array" or "combination array" or grammatical equivalents herein is meant a plurality of individual arrays, as outlined above. Generally the number of individual arrays is set by the size of the microtiter plate used; thus, 96 well, 384 well and 1536 well microtiter plates utilize composite arrays comprising 96, 384 and 1536 individual arrays, although as will be appreciated by those in the art, not each microtiter well need contain an individual array. It should be noted that the composite arrays can comprise individual arrays that are identical, similar or different. That is, in some embodiments, it may be desirable to do the same 2,000 assays on 96 different samples;
alternatively, doing 192,000 experiments on the same sample (i.e. the same sample in each of the 96 wells) may be desirable.
Alternatively, each row or column of the composite array could be the same, for redundancy/quality control. As will be appreciated by those in the art, there are a variety of ways to configure the system.
In addition, the random nature of the arrays may mean that the same population of beads may be added to two different surfaces, resulfing in substantially similar but perhaps not identical arrays.

At least one surface of the substrate is modified to contain discrete, individual sites for later association of microspheres. These sites may comprise physically altered sites, i.e. physical configurations such as wells or small depressions in the substrate that can retain the beads, such that a microsphere can rest in the well, or the use of other forces (magnefic or compressive), or chemically altered or active sites, such as chemically functionalized sites, electrostafically altered sites, hydrophobically/ hydrophilically functionalized sites, spots of adhesive, etc.

The sites may be a pattern, i.e. a regular design or configuration, or randomly distributed. A preferred embodiment utilizes a regular pattern of sites such that the sites may be addressed in the X-Y
coordinate plane. "Pattern" in this sense includes a repeating unit cell, preferably one that allows a high density of beads on the substrate. However, it should be noted that these sites may not be discrete sites. That is, it is possible to use a uniform surface of adhesive or chemical functionalities, for example, that allows the attachment of beads at any position. That is, the surface of the substrate is modified to allow attachment of the microspheres at individual sites, whether or not those sites are configuous or non-contiguous with other sites. Thus, the surface of the substrate may be modified such that discrete sites are formed that can only have a single associated bead, or alternatively, the surface of the substrate is modified and beads may go down anywhere, but they end up at discrete sites.

In a preferred embodiment, the surface of the substrate is modified to contain wells, i.e. depressions in the surface of the substrate. This may be done as is generally known in the art using a variety of 509,13-4 techniques, Including, but not iimited to, photofithography, stamping techniques, molding techniques and microetching techniques. As will be appreciated by those In the art, the technique used will depend on the composition and shape of the substrate.

In a preferred embodiment, physical alterations are made In a surface of the substrate to produce the sites. In a preferred embodiment, the substrate is a fiber optic bundle and the surface of the substrate Is a terminal end of the fiber bundle.
In this embodiment, wells are made in a terminal or distal end of a fiber optic bundle comprising indiyidual flbers.
In this embodiment, the cores of the indivldual fibers are etched, with respect to the cladding, such that small wells or depressions are formed at one end of the fibers. The required depth of the wells wUl depend on the size of the beads to be added to the wells.

Generally in this embodiment, the microspheres are non-covalently associated In the wells, although the wells may additionally be chemically functionalized as is generally described below, cross-linking agents may be used, or a physical barrier may be used, i.e. a film or membrane over the beads.

In a preferred embodiment, the surface of the substrate is modified to contain chemically modified sites, that can be used to attach, either covalently or non-covalently, the microspheres of the invention to the discrete sites or iocations on the substrate. 'Chemicaliy modified sites' in this context includes, but is not iimited to, the addition of a paitem of chemical functional groups induding amino groups, carboxy groups, oxo groups and thiol groups, that can be used to covalently attach microspheres, which generally also contain corresponding reactive func6ional groups; the addidon of a pattem of adhesive that can be used to bind the microspheres (either by prior chemicai functionalization for the addibon of the adhesive or direct addition of the adhesive); the addition of a pattern of charged groups (similar to the chemical fundionaiitles) for the electrostatic attachment of the microspheres, i.e. when the microspheres comprise charged groups opposite to the sites; the addition of a pattern of chemical functional groups that renders the sites differentially hydrophobic or hydrophilic, such that the addition of similarly hydrophobic or hydrophilic microspheres under suitable expedmental cond'dions will result in association of the microspheres to the sites on the basis of hydroaffinity.
For example, the use of hydrophobic sites with hydrophobic beads, in an aqueous system, drives the assodation of the beads preferentiaily onto the sites. As outlined above, "pattem" in this sense Includes the use of a uniform treatment of the surface to allow attachment of the beads at discrete sites, as well as treatment of the surface resuiting In discrete sites. As will be appreciated by those in the art, this may be accomplished In a variety of ways.

In some embodiments, the beads are not associated with a substrate. That Is, the beads are In solution or are not distributed on a pattemed substrate.

In a preferred embodiment, the compositions of the invention further comprise a population of microspheres. By "population" herein is meant a plurality of beads as outlined above for arrays.
Within the population are separate subpopulations, which can be a single microsphere or multiple identical microspheres. That is, in some embodiments, as is more fully outlined below, the array may contain only a single bead for each capture probe; preferred embodiments utilize a plurality of beads of each type.

By "microspheres" or "beads" or "par6cies" or grammatical equivalents herein is meant small discrete parbcies. The composition of the beads will vary, depending on the class of capture probe and the method of synthesis. Suitable bead compositions include those used in peptide, nucleic acid and organic moiety synthesis, including, but not limited to, plastics, ceramics, glass, polystyrene, methylstyrene, acrylic polymers, paramagnetic materials, thoria sol, carbon graphite, titanium dioxide, latex or cross-linked dextrans such as Sepharose, cellulose, nylon, cross-linked micelles and Teflon may all be used. "Microsphere Detection Guide" from Bangs Laboratories, Fishers IN is a helpful guide.

The beads need not be spherical; irregular par6cles may be used. In addition, the beads may be porous, thus increasing the surface area of the bead available for either capture probe attachment or tag attachment. The bead sizes range from nanometers, i.e. 100 nm, to millimeters, i.e. 1 mm, with beads from about 0.2 micron to about 200 microns being preferred, and from about 0.5 to about 5 micron being particularly preferred, although in some embodiments smaller beads may be used.

It should be noted that a key component of the invention is the use of a substrate/bead pairing that allows the association or attachment of the beads at discrete sites on the surface of the substrate, such that the beads do not move during the course of the assay.

Each microsphere comprises a capture probe, although as will be appreciated by those in the art, there may be some microspheres which do not contain a capture probe, depending on the synthetic methods.

Attachment of the nucleic acids may be done in a variety of ways, as will be appreciated by those in the art, including, but not limited to, chemical or affinity capture (for example, including the incorporation of derivatized nucieotides such as AminoLink or biotinylated nucleofides that can then be used to attach the nucleic acid to a surface, as well as affinity capture by hybridization), cross-linking, and electrostatic attachment, etc. In a preferred embodiment, affinity capture is used to attach the nucleic acids to the beads. For example, nucleic acids can be derivatized, for example with one member of a binding pair, and the beads derivatized with the other member of a binding pair. Suitable binding pairs are as described herein for IBUDBL pairs. For example, the nucleic acids may be biotinylated (for example using enzymatic incorporate of biotinylated nucleotides, for by photoactivated cross-linking of biotin). Biotinylated nucleic acids can then be captured on streptavidin-coated beads, as is known in the art. Similarly, other hapten-receptor combinations can be used, such as digoxigenin and anti-digo(igenin antibodies. Alternatively, chemical groups can be added in the form of derivatized nucleotides, that can them be used to add the nucleic acid to the surface.

Preferred attachments are covalent, although even relatively weak interactions (i.e. non-covalent) can be sufficient to attach a nucleic acid to a surface, if there are muitiple sites of attachment per each nucleic acid. Thus, for example, electrostatic interactions can be used for attachment, for example by having beads carrying the opposite charge to the bioactive agent.

Similarly, afflnity capture utilizing hybridizafion can be used to attach nucleic acids to beads. For example, as is known in the art, polyA+RNA is routinely captured by hybridization to oligo-dT beads;
this may include oligo-dT capture followed by a cross-linking step, such as psoralen crosslinking). If the nucleic acids of interest do not contain a polyA tract, one can be attached by polymerization with terminal transferase, or via ligafion of an oligoA linker, as is known in the art.

Alternatively, chemical crosslinking may be done, for example by photoactivated crosslinking of thymidine to reactive groups, as is known in the art.

In general, probes of the present invention are designed to be complementary to a target sequence (either the target sequence of the sample or to other probe sequences, as is described herein), such that hybridization of the target and the probes of the present invention occurs. This complementarily need not be perfect; there may be any number of base pair mismatches that will interfere with hybridization between the target sequence and the single stranded nucleic acids of the present invention. However, if the number of mutations is so great that no hybridization can occur under even the least stringent of hybridizafion conditions, the sequence is not a complementary target sequence.
Thus, by "substantially complementary" herein is meant that the probes are sufficiently complementary to the target sequences to hybridize under the selected reaction conditions.

In a preferred embodiment, each bead comprises a single type of capture probe, although a plurality of individual capture probes are preferably attached to each bead. Similarly, preferred embodiments utilize more than one microsphere containing a unique capture probe; that is, there is redundancy built into the system by the use of subpopulations of microspheres, each microsphere in the subpopulation containing the same capture probe.

As will be appreciated by those in the art, the capture probes may either be synthesized directly on the beads, or they may be made and then attached after synthesis. In a preferred embodiment, linkers are used to attach the capture probes to the beads, to allow both good attachment, sufficient flexibility to allow good interaction with the target molecule, and to avoid undesirable binding reactions.

In a preferred embodiment, the capture probes are synthesized directly on the beads. As is known in the art, many classes of chemical compounds are currently synthesized on solid supports, such as peptides, organic moieties, and nucleic acids. It is a relatively straightforward matter to adjust the current synthetic techniques to use beads.

In a preferred embodiment, the capture probes are synthesized first, and then covalently attached to the beads. As will be appreciated by those in the art, this will be done depending on the composition of the capture probes and the beads. The functionalization of solid support surfaces such as certain polymers with chemically reactive groups such as thiols, amines, carboxyls, etc. is generally known in the art. Accordingly, "blank" microspheres may be used that have surface chemistries that facilitate the attachment of the desired funcfionality by the user. Some examples of these surface chemistries for blank microspheres include, but are not limited to, amino groups including aliphatic and aromatic amines, carboxylic acids, aidehydes, amides, chloromethyl groups, hydrazide, hydroxyl groups, sulfonates and sulfates.

When random arrays are used, an encoding/decoding system must be used. For exampe, when microsphere arrays are used, the beads are generally put onto the substrate randomly; as such there are several ways to correlate the funcfionality on the bead with its location, including the incorporation of unique optical signatures, generally fluorescent dyes, that could be used to identify the chemical functionality on any particular bead. This allows the synthesis of the candidate agents (i.e. compounds such as nucleic acids and antibodies) to be divorced from their placement on an array, i.e. the candidate agents may be synthesized on the beads, and then the beads are randomly distributed on a patterned surface. Since the beads are first coded with an optical signature, this means that the array can later be "decoded", i.e. after the array is made, a correlation of the location of an individual site on the array with the bead or candidate agent at that particular site can be made. This means that the beads may be randomly distributed on the array, a fast and inexpensive process as compared to either the in situ synthesis or spotting techniques of the prior art.

However, the drawback to these methods is that for a large array, the system requires a large number of different optical signatures, which may be difficult or time-consuming to utilize. Accordingly, the present invention provides several improvements over these methods, generally directed to methods of coding and decoding the arrays. That is, as will be appreciated by those in the art, the placement of the capture probes is generally random, and thus a coding/decoding system is required to identify the probe at each location in the array. This may be done in a va(ety of ways, as is more fully outlined below, and generally includes: a) the use a decoding binding ligand (DBL), generally directly labeled, that binds to either the capture probe or to idenfifier binding ligands (IBLs) attached to the beads; b) positional decoding, for example by either targeting the placement of beads (for example by using photoactivatible or photocleavable moieties to allow the selective addition of beads to particular locations), or by using either sub-bundles or selective loading of the sites, as are more fully outlined below; c) selective decoding, wherein only those beads that bind to a target are decoded; or d) combinations of any of these. In some cases, as is more fully outlined below, this decoding may occur for all the beads, or only for those that bind a particular target sequence.
Similarly, this may occur either prior to or after addition of a target sequence. In addifion, as outlined herein, the target sequences detected may be either a primary target sequence (e.g. a pafient sample), or a reaction product from one of the methods described herein (e.g. an extended SBE probe, a ligated probe, a cleaved signal probe, etc.).

Once the identity (i.e. the actual agent) and location of each microsphere in the array has been fixed, the array is exposed to samples containing the target sequences, although as outlined below, this can be done prior to or during the analysis as well. The target sequences can hybridize (either directly or indirectly) to the capture probes as is more fully outlined below, and results in a change in the optical signal of a particular bead.

In the present invention, "decoding" does not rely on the use of op6cal signatures, but rather on the use of decoding binding ligands that are added during a decoding step. The decoding binding ligands will bind either to a distinct identifier binding ligand partner that is placed on the beads, or to the capture probe itself. The decoding binding ligands are either directly or indirectly labeled, and thus decoding occurs by detecting the presence of the label. By using pools of decoding binding ligands in a sequential fashion, it is possible to greatly minimize the number of required decoding steps.

In some embodiments, the microspheres may additionally comprise identifier binding ligands for use in certain decoding systems. By "identifier binding ligands" or "IBLs" herein is meant a compound that will specifically bind a corresponding decoder binding ligand (DBL) to facilitate the elucidation of the identity of the capture probe attached to the bead. That is, the IBL and the corresponding DBL form a binding partner pair. By "specifically bind" herein is meant that the IBL
binds its DBL with specificity sufficient to differentiate between the corresponding DBL and other DBLs (that is, DBLs for other IBLs), or other components or contaminants of the system. The binding should be sufficient to remain bound under the conditions of the decoding step, including wash steps to remove non-specific binding.
In some embodiments, for example when the IBLs and corresponding DBLs are proteins or nucleic acids, the dissociation constants of the IBL to its DBL will be less than about 10-4-10-6 M-', with less than about 10-5 to 10-9 M-' being preferred and less than about 10-' -10-9 M-' being particularly preferred.

IBL-DBL binding pairs are known or can be readily found using known techniques. For example, when the IBL is a protein, the DBLs include proteins (particularly including anfibodies or fragments thereof (FAbs, etc.)) or small molecules, or vice versa (the IBL is an antibody and the DBL is a protein). Metal ion- metal ion ligands or chelators pairs are also useful. Antigen-antibody pairs, enzymes and substrates or inhibitors, other protein-protein interacting pairs, receptor-ligands, complementary 50,913-4 nucleic acids, and carbohydrates and their binding partners are also suitabie binding pairs. Nudeic acid - nucleic acid binding proteins pairs are also useful. Similarly, as is generally described In U.S.
Patents 5,270,163, 5,475,096, 5,567,588, 5,595,877, 5,637,459, 5,683,867,5,705,337, and related patents, nucleic acid 'aptamers' can be developed for binding to virtually any targeY such an aptamer4arget pair can be used as the IBL-DBL
pair. Similarly, there is a wide body of literature relating to the development of binding pairs based on combinatorlai chemistry methods.

In a preferred embodiment, the IBL is a molecule whose color or luminescence properbes change in the presence of a selectively-binding DBL. For example, the IBL may be a fluorescent pH indicator whose emission intensity changes with pH. Simiiarly, the IBL may be a fluorescent Ion indicator, whose emission properties change with Ion concentration.

Afternatively, the IBL is a molecule whose color or luminescence properties change in the presence of various solvents. For example, the IBL may be a fluorescent molecule such as an ethidium salt whose fluorescence intensity increases in hydrophobic environments. Similarly, the IBL may be a derivative of fluorescein whose color changes between aqueous and nonpolar solvents.

In one embodiment, the DBL may be attached to a bead, i.e. a'decoder bead', that may carry a label such as a fluorophore.

In a preferred embodiment, the IBL-DBL pair comprise substantially complementary single-stranded nucleic acids. In this embodiment, the binding ligands can be referred to as'tidentifler probes' and "decoder probes". Generally, the identifier and decoder probes range from about 4 basepairs In length to about 1000, with from about 6 to about 100 being preferred, and from about 8 to about 40 being particutariy preferred. What is important is that the probes are long enough to be spedfic, i.e. to distinguish between different IBL-DBL pairs, yet short enough to allow both a) dissociation, if necessary, under suitabie experimental condidons, and b) effldent hybrid"tzation.

In a preferred embodiment, as is more fully outiined below, the IBLs do not bind to DBLs. Rather, the IBLs are used as identifier moieties ("IMs') that are identified directly, for example through the use of mass spectroscopy.

Alternatively, in a preferred embodiment, the IBL and the capture probe are the same moiety; thus, for example, as outlined herein, particularly when no optical signatures are used, the capture probe can serve as both the iden6fier and the agent. For example, in the case of nuclsic acids, the bead-bound probe (which serves as the capture probe) can also bind decoder probes, to identifythe sequence of the probe on the bead. Thus, in this embodiment, the DBLs bind to the capture probes.

In a preferred embodiment, the microspheres may contain an opiical signature.
That is, previous work had each subpopuiation of microspheres comprising a unique optjcal signature or opflcai tag that Is used to identify the unique capture probe of that subpopuiation of microspheres; that Is, decoding utllizes optlcai properties of the beads such that a bead comprising the unique optical signature may be distinguished from beads at other iocations with different opticai signatures. Thus the previous work assigned each capture probe a unique optical signature such that any microspheres comprising that capture probe are Identifiable on the basis of the signature. These opticai signatures comprised dyes, usually chromophores or fluorophores, that were entrapped or attached to the beads themselves. biversity of opticai signatures utiiized different fluorochromes, different ratios of mbctures of fluorochromes, and different concentrations (ntensitles) of fluorochromes.

In a preferred embodiment, the present invention does not rely solely on the use of optical propertles to decode the arrays. However, as will be appreciated by those In the art, It is possible in some embodiments to utilize optical signatures as an additional coding method, In conjundion with the present system. Thus, for example, as is more fully outlined below, the size of the array may be effectively increased while using a single set of decoding moieties In several ways, one of which is the use of optical signatures one some beads. Thus, for example, using one "set"
of decoding molecules, the use of two popuiatjons of beads, one with an optical signature and one without, allows the effedive doubling of the array size. The use of muitiple opticai signatures similarly increases the possible size of the array.

In a preferred embodiment, each subpopulation of beads comprises a piuraiity of different IBLs. By using a piuraiiiy of different IBLs to encode each capture probe, the number of possible unique codes is substantially increased. That is, by using one unique IBL per capture probe, the size of the array will be the number of unique IBLs (assuming no "reuse" occurs, as outlined below).
However, by using a pluraiity of different IBLs per bead, n, the size of the array can be increased to 2", when the presence or absence of each IBL is used as the indicator. For example, the assignment of 101BLs per bead generates a 10 bit binary code, where each bit can be designated as 81" (IBL
is present) or "0" pBL is absent). A 10 bit binary code has 210 possible variants However, as Is more fully discussed below, the size of the array may be further increased if another parameter is Included such as concentraiion or intensity; thus for example, if two different concentrations of the IBL are used, then the array size Increases as 31. Thus, in this embodiment, each individual capture probe In the array is assigned a combination of IBLs, which can be added to the beads prior to the addition of the capture probe, after, or during the synthesis of the capture probe, i.e. simultaneous addition of IBLs and capture probe components.

Alternatively, the combination of different IBLs can be used to elucidate the sequence of the nudeic acid. Thus, for example, using two different IBLs (IBL1 and IBL2), the first position of a nucleic acid can be elucidated: for example, adenosine can be represented by the presence of both IBL1 and IBL2;
thymidine can be represented by the presence of IBL1 but not IBL2, cytosine can be represented by the presence of IBL2 but not IBL1, and guanosine can be represented by the absence of both. The second position of the nucleic acid can be done in a similar manner using IBL3 and IBL4; thus, the presence of IBL1, IBL2, IBL3 and IBL4 gives a sequence of AA; IBL1, IBL2, and IBL3 shows the sequence AT; IBL1, IBL3 and IBL4 gives the sequence TA, etc. The third position utilizes IBL5 and IBL6, etc. In this way, the use of 20 different identifiers can yield a unique code for every possible 10-mer.

In this way, a sort of "bar code" for each sequence can be constructed; the presence or absence of each distinct IBL will allow the identification of each capture probe.

In addition, the use of different concentrations or densities of IBLs allows a "reuse" of sorts. If, for example, the bead comprising a first agent has a 1X concentration of IBL, and a second bead comprising a second agent has a 10X concentration of IBL, using saturating concentrations of the corresponding labelled DBL allows the user to distinguish between the two beads.

Once the microspheres comprising the capture probes are generated, they are added to the substrate to form an array. It should be noted that while most of the methods described herein add the beads to the substrate prior to the assay, the order of making, using and decoding the array can vary. For example, the array can be made, decoded, and then the assay done.
Alternatively, the array can be made, used in an assay, and then decoded; this may find parbcular use when only a few beads need be decoded. Alternatively, the beads can be added to the assay mixture, i.e.
the sample containing the target sequences, prior to the addition of the beads to the substrate;
after addition and assay, the array may be decoded. This is particularly preferred when the sample comprising the beads is agitated or mixed; this can increase the amount of target sequence bound to the beads per unit time, and thus (in the case of nucleic acid assays) increase the hybridization kinefics. This may find particular use in cases where the concentrafion of target sequence in the sample is low; generally, for low concentrations, long binding times must be used.

In general, the methods of making the arrays and of decoding the arrays is done to maximize the number of different candidate agents that can be uniquely encoded. The compositions of the invention may be made in a variety of ways. In general, the arrays are made by adding a solution or slurry comprising the beads to a surface containing the sites for attachment of the beads. This may be done in a variety of buffers, including aqueous and organic solvents, and mixtures.
The solvent can evaporate, and excess beads are removed.

In a preferred embodiment, when non-covalent methods are used to associate the beads with the array, a novel method of loading the beads onto the array is used. This method comprises exposing the array to a solution of particles (including microspheres and cells) and then applying energy, e.g.
agitating or vibrating the mixture. This results in an array comprising more tightly associated particles, as the agitation is done with sufficient energy to cause weakly-associated beads to fall off (or out, in the case of wells). These sites are then available to bind a different bead.
In this way, beads that exhibit a high affinity for the sites are selected. Arrays made in this way have two main advantages as compared to a more static loading: first of all, a higher percentage of the sites can be filled easily, and secondly, the arrays thus loaded show a substantial decrease in bead loss during assays. Thus, in a preferred embodiment, these methods are used to generate arrays that have at least about 50% of the sites filled, with at least about 75% being preferred, and at least about 90%
being particularly preferred. Similarly, arrays generated in this manner preferably lose less than about 20% of the beads during an assay, with less than about 10% being preferred and less than about 5% being particularly preferred.

In this embodiment, the substrate comprising the surface with the discrete sites is immersed into a solution comprising the parbcies (beads, cells, etc.). The surface may comprise wells, as is described herein, or other types of sites on a patterned surface such that there is a differential affinity for the sites. This differnetial affinity results in a competitive process, such that particles that will associate more tightly are selected. Preferably, the entire surface to be "loaded" with beads is in fluid contact with the solution. This solution is generally a slurry ranging from about 10,000:1 beads:solution (vol:vol) to 1:1. Generally, the solution can comprise any number of reagents, including aqueous buffers, organic solvents, salts, other reagent components, etc. In addition, the solution preferably comprises an excess of beads; that is, there are more beads than sites on the array. Preferred embodiments utilize two-fold to billion-fold excess of beads.

The immersion can mimic the assay conditions; for example, if the array is to be "dipped" from above into a microtiter plate comprising samples, this configuration can be repeated for the loading, thus minimizing the beads that are likely to fall out due to gravity.

Once the surface has been immersed, the substrate, the solution, or both are subjected to a competitive process, whereby the particies with lower affinity can be disassociated from the substrate and replaced by pardcles exhibiting a higher affinity to the site. This competitive process is done by the introduction of energy, in the form of heat, sonication, stirring or mixing, vibrating or agitating the solution or substrate, or both.

A preferred embodiment utilizes agitation or vibration. In general, the amount of manipulation of the substrate is minimized to prevent damage to the array; thus, preferred embodiments utilize the agitafion of the solution rather than the array, although either will work. As will be appreciated by those in the art, this agitation can take on any number of forms, with a preferred embodiment utilizing microtiter plates comprising bead solutions being agitated using microtiter plate shakers.

The agitation proceeds for a period of time suffident to load the array to a desired flll. Depending on the size and concentration of the beads and the size of the array, this tune may range from about 1 second to days, with from about 1 minute to about 24 hours being preferred.

It should be noted that not all sites of an array may comprise a bead; that Is, there may be some sites on the substrate surface which are empty. In addition, there may be some sites that contain more than one bead, aRhough this is not preferred.

In some embodiments, for example when chemical attachment Is done, it is possible to attach the beads in a non-random or ordered way. For example, using photoactivatibie attachment linkers or photoactivatible adhesives or masks, selected sites on the array may be sequentially rendered su'itable for attachment, such that defined populatlons of beads are laid down.

The arrays of the present invention are constructed such that information about the identity of the capture probe is built into the array, such that the random deposidon of the beads in the fiber wells can be 'decoded' to allow identification of the capture probe at all positions.
This may be done In a variety of ways, and either before, during or after the use of the array to detect target molecules.

Thus, after the array is made, it is "decoded' in order to identify the location of one or more of the capture probes, i.e. each subpopulation of beads, on the substrate surface.

In a preferred embodiment, pyrosequencing techniques are used to decode the array.
In a preferred embodiment, a selective decoding system is used. In this case, only those microspheres exhibiting a change in the optical signal as a result of the binding of a target sequence are decoded. This is commonly done when the number of "hits', i.e. the number of sites to decode, is generally low. That is, the array is first scanned under experimental conditions In the absence of the target sequences. The sample containing the target sequences Is added, and only those locations exhibiting a change In the opticai signal are decoded. For example, the beads at e'ither the posiffve or negative signal locations may be either selectively tagged or released from the array (for example through the use of photocleavabie linkers), and subsequently sorted or entiched In a fluorescence-activated cell sorter (FACS). That is, either all the negative beads are released, and then the positnre beads are either released or analyzed in situ, or alternativeiy all the posiitivves are released and analyzed. Altematively, the labels may comprise halogenated aromatic compounds, and detectlon of the label is done using for example gas chromatography, chemical tags, Isotopic tags mass spectral tags.

As will be appreciated by those in the art, this may also be done in systems where the array is not decoded; i.e. there need not ever be a correlation of bead composition with location. In this embodiment, the beads are loaded on the array, and the assay is run. The "positives", i.e. those beads displaying a change in the opfical signal as is more fully outlined below, are then "marked" to distinguish or separate them from the "negative" beads. This can be done in several ways, preferably using fiber optic arrays. In a preferred embodiment, each bead contains a fluorescent dye. After the assay and the identificafion of the "positives" or "active beads", light is shown down either only the positive fibers or only the negative fibers, generally in the presence of a light-activated reagent (typically dissolved oxygen). In the former case, all the active beads are photobleached. Thus, upon non-selective release of all the beads with subsequent sorbng, for example using a fluorescence activated cell sorter (FACS) machine, the non-fluorescent active beads can be sorted from the fluorescent negative beads. Alternatively, when light is shown down the negative fibers, all the negatives are non-fluorescent and the the postives are fluorescent, and sorting can proceed. The characterization of the attached capture probe may be done directly, for example using mass spectroscopy.

Alternatively, the identification may occur through the use of identifier moieties ("IMs"), which are similar to IBLs but need not necessarily bind to DBLs. That is, rather than elucidate the structure of the capture probe directly, the composition of the IMs may serve as the identifier. Thus, for example, a specific combination of IMs can serve to code the bead, and be used to identify the agent on the bead upon release from the bead followed by subsequent analysis, for example using a gas chromatograph or mass spectroscope.

Alternatively, rather than having each bead contain a fluorescent dye, each bead comprises a non-fluorescent precursor to a fluorescent dye. For example, using photocleavable protecting groups, such as certain ortho-nitrobenzyl groups, on a fluorescent molecule, photoactivation of the fluorochrome can be done. After the assay, light is shown down again either the "positive" or the "negative" fibers, to distinquish these populations. The illuminated precursors are then chemically converted to a fluorescent dye. All the beads are then released from the array, with sorBng, to form populafions of fluorescent and non-fluorescent beads (either the positives and the negatives or vice versa).

In an alternate preferred embodiment, the sites of attachment of the beads (for example the wells) include a photopolymerizable reagent, or the photopolymerizable agent is added to the assembled array. After the test assay is run, light is shown down again either the "positive" or the "negative"
fibers, to distinquish these populations. As a result of the irradiation, either all the positives or all the negatives are polymerized and trapped or bound to the sites, while the other population of beads can be released from the array.

In a preferred embodiment, the location of every capture probe is determined using decoder binding ligands (DBLs). As outlined above, DBLs are binding ligands that will either bind to identifier binding ligands, if present, or to the capture probes themselves, preferably when the capture probe is a nucleic acid or protein.

In a preferred embodiment, as outlined above, the DBL binds to the IBL.

In a preferred embodiment, the capture probes are single-stranded nucleic acids and the DBL is a substantially complementary single-stranded nucleic acid that binds (hybridizes) to the capture probe, termed a decoder probe herein. A decoder probe that is substantially complementary to each candidate probe is made and used to decode the array. In this embodiment, the candidate probes and the decoder probes should be of sufficient length (and the decoding step run under suitable conditions) to allow specificity; i.e. each candidate probe binds to its corresponding decoder probe with sufficient specificity to allow the distinction of each candidate probe.

In a preferred embodiment, the DBLs are either directly or indirectly labeled.
In a preferred embodiment, the DBL is directly labeled, that is, the DBL comprises a label.
In an alternate embodiment, the DBL is indirectly labeled; that is, a labeling binding ligand (LBL) that will bind to the DBL is used. In this embodiment, the labeling binding ligand-DBL pair can be as described above for IBL-DBL pairs.

Accordingly, the identification of the location of the individual beads (or subpopulations of beads) is done using one or more decoding steps comprising a binding between the labeled DBL and either the IBL or the capture probe (i.e. a hybridizafion between the candidate probe and the decoder probe when the capture probe is a nucleic acid). After decoding, the DBLs can be removed and the array can be used; however, in some circumstances, for example when the DBL binds to an IBL and not to the capture probe, the removal of the DBL is not required (although it may be desirable in some circumstances). In addition, as outlined herein, decoding may be done either before the array is used to in an assay, during the assay, or after the assay.

In one embodiment, a single decoding step is done. In this embodiment, each DBL is labeled with a unique label, such that the the number of unique tags is equal to or greater than the number of capture probes (although in some cases, "reuse" of the unique labels can be done, as described herein;
similarly, minor variants of candidate probes can share the same decoder, if the variants are encoded in another dimension, i.e. in the bead size or label). For each capture probe or IBL, a DBL is made that will specifically bind to it and contains a unique tag, for example one or more fluorochromes.
Thus, the identity of each DBL, both its composition (i.e. its sequence when it is a nucleic acid) and its label, is known. Then, by adding the DBLs to the array containing the capture probes under conditions which allow the formation of complexes (termed hybridization complexes when the components are nucleic acids) between the DBLs and either the capture probes or the IBLs, the location of each DBL
can be elucidated. This allows the identification of the location of each capture probe; the random array has been decoded. The DBLs can then be removed, if necessary, and the target sample applied.

In a preferred embodiment, the number of unique labels is less than the number of unique capture probes, and thus a sequential series of decoding steps are used. In this embodiment, decoder probes are divided into n sets for decoding. The number of sets corresponds to the number of unique tags.
Each decoder probe is labeled in n separate reactions with n disfinct tags.
All the decoder probes share the same n tags. The decoder probes are pooled so that each pool contains only one of the n tag versions of each decoder, and no two decoder probes have the same sequence of tags across all the pools. The number of pools required for this to be true is determined by the number of decoder probes and the n. Hybridization of each pool to the array generates a signal at every address. The sequential hybridization of each pool in turn will generate a unique, sequence-specific code for each candidate probe. This idenfifies the candidate probe at each address in the array. For example, if four tags are used, then 4 X n sequential hybridizations can ideally distinguish 4"
sequences, although in some cases more steps may be required. After the hybridization of each pool, the hybrids are denatured and the decoder probes removed, so that the probes are rendered single-stranded for the next hybridization (although it is also possible to hybridize limiting amounts of target so that the available probe is not saturated. Sequential hybridizations can be carried out and analyzed by subtracting pre-existing signal from the previous hybridization).

An example is illustrative. Assuming an array of 16 probe nucleic acids (numbers 1-16), and four unique tags (four different fluors, for example; labels A-D). Decoder probes 1-16 are made that correspond to the probes on the beads. The first step is to label decoder probes 1-4 with tag A, decoder probes 5-8 with tag B, decoder probes 9-12 with tag C, and decoder probes 13-16 with tag D.
The probes are mixed and the pool is contacted with the array containing the beads with the attached candidate probes. The locafion of each tag (and thus each decoder and candidate probe pair) is then determined. The first set of decoder probes are then removed. A second set is added, but this time, decoder probes 1, 5, 9 and 13 are labeled with tag A, decoder probes 2, 6, 10 and 14 are labeled with tag B, decoder probes 3, 7, 11 and 15 are labeled with tag C, and decoder probes 4, 8, 12 and 16 are labeled with tag D. Thus, those beads that contained tag A in both decoding steps contain candidate probe 1; tag A in the first decoding step and tag B in the second decoding step contain candidate probe 2; tag A in the first decoding step and tag C in the second step contain candidate probe 3; etc.
In one embodiment, the decoder probes are labeled in situ; that is, they need not be labeled prior to the decoding reaction. In this embodiment, the incoming decoder probe is shorter than the candidate probe, creating a 5' "overhang" on the decoding probe. The addition of labeled ddNTPs (each labeled with a unique tag) and a polymerase will allow the addition of the tags in a sequence specific manner, thus creating a sequence-specific pattern of signals. Similarly, other modificafions can be done, including ligation, etc.

In addition, since the size of the array will be set by the number of unique decoding binding ligands, it is possible to "reuse" a set of unique DBLs to allow for a greater number of test sites. This may be done in several ways; for example, by using some subpopulations that comprise optical signatures.
Similarly, the use of a positional coding scheme within an array; different sub-bundles may reuse the set of DBLs. Similarly, one embodiment utilizes bead size as a coding modality, thus allowing the reuse of the set of unique DBLs for each bead size. Alternatively, sequen6al partial loading of arrays with beads can also allow the reuse of DBLs. Furthermore, "code sharing" can occur as well.

In a preferred embodiment, the DBLs may be reused by having some subpopulations of beads comprise opfical signatures. In a preferred embodiment, the opfical signature is generally a mixture of reporter dyes, preferably fluoroscent. By varying both the composition of the mixture (i.e. the ratio of one dye to another) and the concentrafion of the dye (leading to differences-in signal intensity), matrices of unique optical signatures may be generated. This may be done by covalently attaching the dyes to the surface of the beads, or alternatively, by entrapping the dye within the bead.

In a preferred embodiment, the encoding can be accomplished in a ratio of at least two dyes, although more encoding dimensions may be added in the size of the beads, for example.
In addition, the labels are disfinguishable from one another; thus two different labels may comprise different molecules (i.e.
two different fluors) or, alternatively, one label at two different concentrations or intensity.

In a preferred embodiment, the dyes are covalently attached to the surface of the beads. This may be done as is generally outlined for the attachment of the capture probes, using funcfional groups on the surface of the beads. As will be appreciated by those in the art, these attachments are done to minimize the effect on the dye.

In a preferred embodiment, the dyes are non-covalently associated with the beads, generally by entrapping the dyes in the pores of the beads.

Additionally, encoding in the ratios of the two or more dyes, rather than single dye concentrations, is preferred since it provides insensitivity to the intensity of light used to interrogate the reporter dye's signature and detector sensifivity.

In a preferred embodiment, a spatial or positional coding system is done. In this embodiment, there are sub-bundles or subarrays (i.e. portions of the total array) that are utilized. By analogy with the telephone system, each subarray is an "area code", that can have the same tags (i.e. telephone numbers) of other subarrays, that are separated by virtue of the locafion of the subarray. Thus, for example, the same unique tags can be reused from bundle to bundle. Thus, the use of 50 unique tags in combination with 100 different subarrays can form an array of 5000 different capture probes. In this embodiment, it becomes important to be able to identify one bundle from another; in general, this is done either manually or through the use of marker beads, i.e. beads containing unique tags for each subarray.

In alternative embodiments, additional encoding parameters can be added, such as microsphere size.
For example, the use of different size beads may also allow the reuse of sets of DBLs; that is, it is possible to use microspheres of different sizes to expand the encoding dimensions of the microspheres. Optical fiber arrays can be fabricated containing pixels with different fiber diameters or cross-sections; alternafively, two or more fiber optic bundles, each with different cross-sections of the individual fibers, can be added together to form a larger bundle; or, fiber optic bundles with fiber of the same size cross-sec6ons can be used, but just with different sized beads. With different diameters, the largest wells can be filled with the largest microspheres and then moving onto progressively smaller microspheres in the smaller wells until all size wells are then filled. In this manner, the same dye ratio could be used to encode microspheres of different sizes thereby expanding the number of different oligonucleotide sequences or chemical funcfionalities present in the array. Although outlined for fiber optic substrates, this as well as the other methods outlined herein can be used with other substrates and with other attachment modalities as well.

In a preferred embodiment, the coding and decoding is accomplished by sequential loading of the microspheres into the array. As outlined above for spatial coding, in this embodiment, the optical signatures can be "reused". In this embodiment, the library of microspheres each comprising a different capture probe (or the subpopulations each comprise a different capture probe), is divided into a plurality of sublibraries; for example, depending on the size of the desired array and the number of unique tags, 10 sublibraries each comprising roughly 10% of the total library may be made, with each sublibrary comprising roughly the same unique tags. Then, the first sublibrary is added to the fiber optic bundle comprising the wells, and the location of each capture probe is determined, generally through the use of DBLs. The second sublibrary is then added, and the location of each capture probe is again determined. The signal in this case will comprise the signal from the "first" DBL and the "second" DBL; by comparing the two matrices the location of each bead in each sublibrary can be determined. Similarly, adding the third, fourth, etc. sublibraries sequentially will allow the array to be filled.

In a preferred embodiment, codes can be "shared" in several ways. In a first embodiment, a single code (i.e. IBUDBL pair) can be assigned to two or more agents if the target sequences different sufficiently in their binding strengths. For example, two nucleic acid probes used in an mRNA
quantitation assay can share the same code if the ranges of their hybridizafion signal intensities do not overlap. This can occur, for example, when one of the target sequences is always present at a much higher concentration than the other. Alternatively, the two target sequences might always be present at a similar concentration, but differ in hybridization efficiency.

Alternatively, a single code can be assigned to muitiple agents if the agents are functionally equivalent.
For example, if a set of oligonucleotide probes are designed with the common purpose of detecting the presence of a particular gene, then the probes are functionally equivalent, even though they may differ in sequence. Similarly, an array of this type could be used to detect homologs of known genes.
In this embodiment, each gene is represented by a heterologous set of probes, hybridizing to different regions of the gene (and therefore differing in sequence). The set of probes share a common code. If a homolog is present, it might hybridize to some but not all of the probes.
The level of homology might be indicated by the fraction of probes hybridizing, as well as the average hybridization intensity.
Similarly, mulfiple antibodies to the same protein could all share the same code.

In a preferred embodiment, decoding of self-assembled random arrays is done on the bases of pH
titration. In this embodiment, in addition to capture probes, the beads comprise optical signatures, wherein the opfical signatures are generated by the use of pH-responsive dyes (sometimes referred to herein as "ph dyes") such as fluorophores. This embodiment is similar to that outlined in PCT
US98/05025 and U.S.S.N. 09/151,877, both of which are expressly incorporated by reference, except that the dyes used in the present ivention exhibits changes in fluorescence intensity (or other properties) when the solution pH is adjusted from below the pKa to above the pKa (or vice versa). In a preferred embodiment, a set of pH dyes are used, each with a different pKa, preferably separated by at least 0.5 pH units. Preferred embodiments utilize a pH dye set of pKa's of 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, 5.0, 5.5, 6.0, 6.5, 7.0, 7.5, 8.0, 8.5, 9.0, 9.5, 10.0, 10.5, 11, and 11.5.
Each bead can contain any subset of the pH dyes, and in this way a unique code for the capture probe is generated. Thus, the decoding of an array is achieved by titrating the array from pH 1 to pH 13, and measuring the fluorescence signal from each bead as a function of solution pH.

Thus, the present invention provides array composifions comprising a substrate with a surface comprising discrete sites. A population of microspheres is distributed on the sites, and the population comprises at least a first and a second subpopulation. Each subpopula6on comprises a capture probe, and, in addition, at least one optical dye with a given pKa. The pKas of the different optical dyes are different.

In a preferred embodiment, "random" decoding probes can be made. By sequential hybridizations or the use of multiple labels, as is outlined above, a unique hybridization pattern can be generated for each sensor element. This allows all the beads representing a given clone to be identified as belonging to the same group. In general, this is done by using random or par6ally degenerate decoding probes, that bind in a sequence-dependent but not highly sequence-specific manner. The process can be repeated a number of times, each time using a different labeling entity, to generate a different pattern of singais based on quasi-specific interactions. In this way, a unique optical signature is eventually built up for each sensor element. By applying pattern recognition or clustering algorithms to the optical signatures, the beads can be grouped into sets that share the same signature (i.e. carry the same probes).

In order to identify the actual sequence of the clone itself, additional procedures are required; for example, direct sequencing can be done, or an ordered array containing the clones, such as a spotted cDNA array, to generate a "key" that links a hybridizafion pattern to a specific clone.

Alternafively, clone arrays can be decoded using binary decoding with vector tags. For example, partially randomized oligos are cloned into a nucleic acid vector (e.g.
plasmid, phage, etc.). Each oligonucleotide sequence consists of a subset of a limited set of sequences.
For example, if the limites set comprises 10 sequences, each oligonucleofide may have some subset (or all of the 10) sequences. Thus each of the 10 sequences can be present or absent in the oligonucleofide.
Therefore, there are 210 or 1,024 possible combinafions. The sequences may overlap, and minor variants can also be represented (e.g. A, C, T and G substitufions) to increase the number of possible combinafions. A nucleic acid library is cloned into a vector containing the random code sequences.
Alternatively, other methods such as PCR can be used to add the tags. In this way it is possible to use a small number of oligo decoding probes to decode an array of clones.

As will be appreciated by those in the art, the systems of the invention may take on a large number of different configurafions, as is generally depicted in the Figures. In general, there are three types of systems that can be used: (1) "non-sandwich" systems (also referred to herein as "direct" detecfion) in which the target sequence itself is labeled with detectable labels (again, either because the primers comprise labels or due to the incorporafion of labels into the newly synthesized strand); (2) systems in which label probes directly bind to the target analytes; and (3) systems in which label probes are indirectly bound to the target sequences, for example through the use of amplifier probes.

Detecfion of the reacfions of the invenfion, including the direct detecfion of products and indirect detection ufilizing label probes (i.e. sandwich assays), is preferably done by detecfing assay complexes comprising detectable labels, which can be attached to the assay complex in a variety of ways, as is more fully described below.

Once the target sequence has preferably been anchored to the array, an amplifier probe is hybridized to the target sequence, either directly, or through the use of one or more label extender probes, which serves to allow "generic" amplifier probes to be made. As for all the steps outlined herein, this may be done simultaneously with capturing, or sequentially. Preferably, the amplifier probe contains a multiplicity of amplificafion sequences, although in some embodiments, as described below, the amplifier probe may contain only a single amplificafion sequence, or at least two amplificafion sequences. The amplifier probe may take on a number of different forms; either a branched conformation, a dendrimer conformafion, or a linear "string" of ampliflcafion sequences. Label probes comprising detectable labels (preferably but not required to be fluorophores) then hybridize to the amplificafion sequences (or in some cases the label probes hybridize directly to the target sequence), and the labels detected, as is more fully outlined below.

Accordingly, the present invenfion provides compositions comprising an amplifier probe. By "amplifier probe" or "nucleic acid mulfimer" or "amplificafion mulfimer" or grammafical equivalents herein is meant a nucleic acid probe that is used to facilitate signal amplificadon.
Amplifier probes comprise at least a first single-stranded nucleic acid probe sequence, as defined below, and at least one single-stranded nucleic acid amplificafion sequence, with a mulfiplicity of amplificafion sequences being preferred.

Amplifier probes comprise a first probe sequence that is used, either directly or indirectly, to hybridize to the target sequence. That is, the amplifier probe itself may have a first probe sequence that is substanfially complementary to the target sequence, or it has a first probe sequence that is substanfially complementary to a portion of an addifional probe, in this case called a label extender probe, that has a first portion that is substanfially complementary to the target sequence. In a preferred embodiment, the first probe sequence of the amplifier probe is substanfially complementary to the target sequence.

In general, as for all the probes herein, the first probe sequence is of a length sufficient to give specificity and stability. Thus generally, the probe sequences of the invenfion that are designed to hybridize to another nucleic acid (i.e. probe sequences, amplification sequences, por6ons or domains of larger probes) are at least about 5 nucleosides long, with at least about 10 being preferred and at least about 15 being especially preferred.

In a preferred embodiment, several different amplifier probes are used, each with first probe sequences that will hybridize to a different portion of the target sequence.
That is, there is more than one level of amplificafion; the amplifier probe provides an amplificafion of signal due to a muifiplicity of labelling events, and several different amplifier probes, each with this muifiplicity of labels, for each target sequence is used. Thus, preferred embodiments ufilize at least two different pools of amplifier probes, each pool having a different probe sequence for hybridizafion to different portions of the target sequence; the only real limitafion on the number of different amplifier probes will be the length of the original target sequence. In addifion, it is also possible that the different amplifier probes contain different amplificafion sequences, although this is generally not preferred.

In a preferred embodiment, the amplifier probe does not hybridize to the sample target sequence directly, but instead hybridizes to a first portion of a label extender probe.
This is particularly useful to allow the use of "generic" amplifier probes, that is, amplifier probes that can be used with a variety of different targets. This may be desirable since several of the amplifier probes require special synthesis techniques. Thus, the addition of a relafively short probe as a label extender probe is preferred. Thus, the first probe sequence of the amplifier probe is substantially complementary to a first portion or domain of a first label extender single-stranded nucleic acid probe. The label extender probe also contains a second porfion or domain that is substantially complementary to a portion of the target sequence. Both of these portions are preferably at least about 10 to about 50 nucleotides in length, with a range of about 15 to about 30 being preferred. The terms "first" and "second" are not meant to confer an orientation of the sequences with respect to the 5'-3' orientation of the target or probe sequences. For example, assuming a 5'-3' orientation of the complementary target sequence, the first por6on may be located either 5' to the second portion, or 3' to the second portion. For convenience herein, the order of probe sequences are generally shown from left to right.

In a preferred embodiment, more than one label extender probe-amplifier probe pair may be used, that is, n is more than 1. That is, a plurality of label extender probes may be used, each with a porfion that is substantially complementary to a different porfion of the target sequence;
this can serve as another level of amplification. Thus, a preferred embodiment utilizes pools of at least two label extender probes, with the upper limit being set by the length of the target sequence.

In a preferred embodiment, more than one label extender probe is used with a single amplifier probe to reduce non-specific binding, as is generally outlined in U.S. Patent No.
5,681,697, incorporated by reference herein. In this embodiment, a first porfion of the first label extender probe hybridizes to a first por6on of the target sequence, and the second portion of the first label extender probe hybridizes to a first probe sequence of the amplifier probe. A first por6on of the second label extender probe hybridizes to a second portion of the target sequence, and the second portion of the second label extender probe hybridizes to a second probe sequence of the amplifier probe.
These form structures sometimes referred to as "cruciform" structures or configurations, and are generally done to confer stability when large branched or dendrimeric amplifier probes are used.

In addition, as will be appreciated by those in the art, the label extender probes may interact with a preamplifier probe, described below, rather than the amplifier probe directly.

Similarly, as outlined above, a preferred embodiment utilizes several different amplifier probes, each with first probe sequences that will hybridize to a different portion of the label extender probe. In addifion, as outlined above, it is also possible that the different amplifier probes contain different amplification sequences, although this is generally not preferred.

In addition to the first probe sequence, the amplifier probe also comprises at least one amplification sequence. An "amplification sequence" or "amplification segment" or grammatical equivalents herein is meant a sequence that is used, either directly or indirectly, to bind to a first porbon of a label probe as is more fully described below. Preferably, the amplifler probe comprises a multlplicily of amplification sequences, with from about 3 to about 1000 being preferred, from about 10 to about 100 being particularly preferred, and about 50 being especially preferred. In some cases, for example when linear amplifier probes are used, from 1 to about 20 is preferred with from about 5 to about 10 being particularly preferred.

The ampi'ification sequences may be linked to each other in a variety of ways, as will be appreciated by those in the art. They may be covalently linked directly to each other, or to intervening sequences or chemical moieties, through nucleic acid linkages such as phosphodiester bonds, PNA bonds, etc., or through interposed linking agents such amino acid, carbohydrate or,polyol bridges, or through other cross-linking agents or binding partners. The site(s) of linkage may be at the ends of a segment, and/or at one or more Internal nucieotides in the strand. In a preferred embodiment, the amplification sequences are attached via nucleic acid linkages.

In a preferred embodiment, branched amplifier probes are used, as are generaily described in U.S.
Patent No. 5,124,246. Branched amplifier probes may take on "fork-like' or "comb-like' conformations. 'Fork-like' branched amplifier probes generally have three or more oligonucleotide segments emanating from a point of origin to form a branched structure. The point of origin may be another nucleotide segment or a multifunctional molecule to whcih at least three segments can be covalently or tightiy bound. "Comb-like" branched amplifier probes have a linear backbone with a mulfiplicity of sidechain oligonucleofides extending from the backbone. In either conformation, the pendant segments will normally depend from a modified nucleotide or other organic moiety having the appropriate func6onal groups for attachment of oGgonucleotides. Furthermore, in either conformation, a large number of amplification sequences are available for binding, either directly or indirectly. to detection probes. In general, these structures are made as is known in the art, using modified muitifunctionai nucieotides, as is described in U.S. Patent Nos.
5,635,352 and 5,124,246, among others.

In a preferred embodiment, dendrimer amplifier probes are used, as are generally described in U.S.
Patent No. 5,175,270. Dendrimeric ampfifier probes have amplification sequences that are attached via hybridization, and thus have portions of double-stranded nucleic acid as a component of their structure. The outer surface of the dendrimer amplifier probe ha's a multiplicity of amplification sequences.

In a preferred embodiment, linear amplifier probes are used, that have Individual ampiiflcation sequences linked end-to-end either directly or with short intervening sequences to form a polymer. As with the other amplifier configurations, there may be additional sequences or moieties between the amplification sequences. In one embodiment, the linear amplifier probe has a single amplification sequence.

In addition, the amplifier probe may be totally linear, totally branched, totally dendrimeric, or any combination thereof.

The amplificafion sequences of the amplifier probe are used, either directly or indirectly, to bind to a label probe to allow detection. In a preferred embodiment, the amplification sequences of the amplifier probe are substantially complementary to a first por6on of a label probe. Alternatively, amplifier extender probes are used, that have a first portion that binds to the amplification sequence and a second por6on that binds to the first portion of the label probe.

In addition, the compositions of the invention may include "preamplifier"
molecules, which serves a bridging moiety between the label extender molecules and the amplifier probes.
In this way, more amplifier and thus more labels are ultimately bound to the detection probes.
Preamplifier molecules may be either linear or branched, and typically contain in the range of about 30-3000 nucleotides.
Thus, label probes are either substantially complementary to an amplification sequence or to a portion of the target sequence.

Detecfion of the nucleic acid reacfions of the invention, including the direct detection of genotyping products and indirect detection utilizing label probes (i.e. sandwich assays), is done by detecfing assay complexes comprising labels.

In a preferred embodiment, several levels of redundancy are built into the arrays of the invention.
Building redundancy into an array gives several significant advantages, including the ability to make quantitative estimates of confidence about the data and signficant increases in sensitivity. Thus, preferred embodiments utilize array redundancy. As will be appreciated by those in the art, there are at least two types of redundancy that can be built into an array: the use of muifiple identical sensor elements (termed herein "sensor redundancy"), and the use of multiple sensor elements directed to the same target analyte, but comprising different chemical functionalities (termed herein "target redundancy"). For example, for the detection of nucleic acids, sensor redundancy utilizes of a plurality of sensor elements such as beads comprising identical binding ligands such as probes. Target redundancy utilizes sensor elements with different probes to the same target:
one probe may span the first 25 bases of the target, a second probe may span the second 25 bases of the target, etc. By building in either or both of these types of redundancy into an array, significant benefits are obtained.
For example, a variety of statistical mathematical analyses may be done.

In addition, while this is generally described herein for bead arrays, as will be appreciated by those in the art, this techniques can be used for any type of arrays designed to detect target analytes.
Furthermore, while these techniques are generally described for nucleic acid systems, these techniques are useful in the detection of other binding ligand/target analyte systems as well.

In a preferred embodiment, sensor redundancy is used. In this embodiment, a plurality of sensor elements, e.g. beads, comprising identical bioactive agents are used. That is, each subpopulation comprises a plurality of beads comprising identical bioactive agents (e.g.
binding ligands). By using a number of identical sensor elements for a given array, the optical signal from each sensor element can be combined and any number of stafistical analyses run, as outlined below.
This can be done for a variety of reasons. For example, in fime varying measurements, redundancy can significantly reduce the noise in the system. For non-fime based measurements, redundancy can significantly increase the confidence of the data.

In a preferred embodiment, a plurality of identical sensor elements are used.
As will be appreciated by those in the art, the number of identical sensor elements will vary with the application and use of the sensor array. In general, anywhere from 2 to thousands may be used, with from 2 to 100 being preferred, 2 to 50 being particularly preferred and from 5 to 20 being especially preferred. In general, preliminary results indicate that roughly 10 beads gives a sufficient advantage, although for some applications, more identical sensor elements can be used.

Once obtained, the optical response signals from a plurality of sensor beads within each bead subpopulation can be manipulated and analyzed in a wide variety of ways, including baseline adjustment, averaging, standard deviation analysis, distribufion and cluster analysis, confidence interval analysis, mean tesfing, etc.

In a preferred embodiment, the first manipulation of the optical response signals is an optional baseline adjustment. In a typical procedure, the standardized optical responses are adjusted to start at a value of 0.0 by subtracting the integer 1.0 from all data points. Doing this allows the baseline-loop data to remain at zero even when summed together and the random response signal noise is canceled out. When the sample is a fluid, the fluid pulse-loop temporal region, however, frequently exhibits a characteristic change in response, either positive, negative or neutral, prior to the sample pulse and often requires a baseline adjustment to overcome noise associated with drift in the first few data points due to charge buildup in the CCD camera. If no drift is present, typically the baseline from the first data point for each bead sensor is subtracted from all the response data for the same bead. If drift is observed, the average baseline from the first ten data points for each bead sensor is substracted from the all the response data for the same bead. By applying this baseline adjustment, when multiple bead responses are added together they can be amplified while the baseline remains at zero. Since all beads respond at the same time to the sample (e.g. the sample pulse), they all see the pulse at the exact same time and there is no registering or adjusfing needed for overlaying their responses. In addition, other types of baseline adjustment may be done, depending on the requirements and output of the system used.

Once the baseline has been adjusted, a number of possible statisgcal analyses may be run to generate known statistical parameters. Analyses based on redundancy are known and generaily described In texts such as Freund and Walpole, Mathematical Sta6stlcs, Prentice Hall, Inc. Nsw Jersey, 1980.

In a preferred embodiment, signal summing is done by simply adding the intensity values of all responses at each time point, generating a new temporal response comprised of the sum of all bead responses. These values can be baseline-adjusted or raw. As for all the analyses described herein, signal summing can be performed in real time or during post-data acquisidon data reduction and analysis. In one embodiment, signal summing is performed with a commerdal spreadsheet program (Excel, Microsoft, Redmond, WA) after opticai response data is collected.

In a preferred embodiment cummulative response data is generated by simply adding all data points in successive time intervals. This final column, comprised of the sum of all data points at a partlcular time interval, may then be compared or plotted with the individual bead responses to determine the extent of signal enhancement or Improved signal-to-noise raiios.

In a preferred embodiment, the mean of the subpopulation (i.e. the plurality of identicai beads) Is determined, using the well known Equation 1:
Equation I

In some embodiments, the subpopulatlon may be redefined to exclude some beads if necessary (for example for obvious ouUiers, as discussed below).

In a preferred embodiment, the standard deviation of the subpopuiaton can be determined, generally using Equation 2 (for the entire subpopulation) and Equation 3 (for less than the entire subpopulation):
Equal3on 2 a=

Equation 3 s= ~~~2 As for the mean, the subpopulation may be redefined to exclude some beads if necessary (for example for obvious outliers, as discussed below).

In a preferred embodiment, stabstcal analyses are done to evaluate whether a particular data point has statistical validity within a subpopulafion by using techniques including, but not limited to, t distribufion and cluster analysis. This may be done to stafistically discard outliers that may othennrise skew the result and increase the signal-to-noise ratio of any particular experiment. This may be done using Equation 4:
Equation 4 t=
~S~2-2 In a preferred embodiment, the quality of the data is evaluated using confidence intervals, as is known in the art. Confidence intervals can be used to facilitate more comprehensive data processing to measure the statistical validity of a result.

In a preferred embodiment, statistical parameters of a subpopula6on of beads are used to do hypothesis testing. One applicafion is tests concerning means, also called mean tesfing. In this application, stabstcal evaluation is done to determine whether two subpopulations are different. For example, one sample could be compared with another sample for each subpopulation within an array to determine if the variation is statisfically significant.

In addition, mean testing can also be used to differenfiate two different assays that share the same code. If the two assays give results that are statistically distinct from each other, then the subpopulations that share a common code can be distinguished from each other on the basis of the assay and the mean test, shown below in Equation 5:
Equation 5 z 61+62 Furthermore, analyzing the distribution of individual members of a subpopulation of sensor elements may be done. For example, a subpopulation distribution can be evaluated to determine whether the distribution is binomial, Poisson, hypergeometric, etc.

In addition to the sensor redundancy, a preferred embodiment utilizes a plurality of sensor elements that are directed to a single target analyte but yet are not identical. For example, a single target nucleic acid analyte may have two or more sensor elements each comprising a different probe. This adds a level of confidence as non-specific binding interactions can be statistically minimized. When nucleic acid target analytes are to be evaluated, the redundant nucleic acid probes may be overlapping, adjacent, or spatially separated. However, it is preferred that two probes do not compete for a single binding site, so adjacent or separated probes are preferred.
Similarly, when proteinaceous target analytes are to be evaluated, preferred embodiments utilize bioactive agent binding agents that bind to different parts of the target. For example, when antibodies (or antibody fragments) are used as bioactive agents for the binding of target proteins, preferred embodiments utilize antibodies to different epitopes.

In this embodiment, a plurality of different sensor elements may be used, with from about 2 to about being preferred, and from about 2 to about 10 being especially preferred, and from 2 to about 5 being particularly preferred, including 2, 3, 4 or 5. Howeve, as above, more may also be used, 15 depending on the application.

As above, any number of statisticat analyses may be run on the data from target redundant sensors.
One benefit of the sensor element summing (referred to herein as "bead summing" when beads are used), is the increase in sensitivity that can occur.

In addition, the present invention is directed to the use of adapter sequences to assemble arrays 20 comprising target analytes. including non-nucleic acid target analytes. By "target analyte" or "analyte"
or grammatical equivalents herein is meant any molecule, compound or particle to be detected. As outlined below, target analytes preferably bind to binding ligands, as is more fully described below. As will be appreciated by those in the art, a large number of analytes may be detected using the present methods; basically, any target analyte for which a binding ligand, described below, may be made may be detected using the methods of the invention.

Suitable analytes include organic and inorganic molecules, including biomolecules. In a preferred embodiment, the analyte may be an environmental pollutant (including pesticides, insecficides, toxins, etc.); a chemical (including solvents, polymers, organic materials, etc.);
therapeutic molecules (including therapeutic and abused drugs, antibiotics, etc.); biomolecules (including hormones, cytokines, proteins, lipids, carbohydrates, cellular membrane antigens and receptors (neural, hormonal, nutrient, and cell surface receptors) or their ligands, etc); whole cells (including procaryotic (such as pathogenic bacteria) and eukaryotic cells, including mammalian tumor cells); viruses (including retroviruses, herpesviruses, adenoviruses, lentiviruses, etc.); and spores; etc. Particularly preferred analytes are environmental pollutants; nucleic acids; proteins (including enzymes, antibodies, antigens, growth factors, cytokines, etc); therapeutic and abused drugs; cells; and viruses.
In a preferred embodiment, the target analyte is a protein. As will be appreciated by those in the art, there are a large number of possible proteinaceous target analytes that may be detected using the present invention. By "proteins" or grammatical equivalents herein is meant proteins, oligopeptides and peptides, derivatives and analogs, including proteins containing non-naturally occurring amino acids and amino acid analogs, and pepfidomimetic structures. The side chains may be in either the (R) or the (S) configuration. In a preferred embodiment, the amino acids are in the (S) or L-configuration. As discussed below, when the protein is used as a binding ligand, it may be desirable to utilize protein analogs to retard degradation by sample contaminants.

Suitable protein target analytes include, but are not limited to, (1) immunoglobulins, particularly IgEs, IgGs and IgMs, and particularly therapeufically or diagnostically relevant anfibodies, including but not limited to, for example, antibodies to human albumin, apolipoproteins (including apolipoprotein E), human chorionic gonadotropin, corasol, a-fetoprotein, thyroxin, thyroid stimulating hormone (TSH), antithrombin, antibodies to pharmaceuticals (including antieptileptic drugs (phenytoin, primidone, carbariezepin, ethosuximide, valproic acid, and phenobarbitol), cardioactive drugs (digoxin, lidocaine, procainamide, and disopyramide), bronchodilators (theophylline), antibiotics (chloramphenicol, sulfonamides), antidepressants, immunosuppresants, abused drugs (amphetamine, methamphetamine, cannabinoids, cocaine and opiates) and antibodies to any number of viruses (including orthomyxoviruses, (e.g. influenza virus), paramyxoviruses (e.g respiratory syncy6al virus, mumps virus, measles virus), adenoviruses, rhinoviruses, coronaviruses, reoviruses, togaviruses (e.g.
rubella virus), parvoviruses, poxviruses (e.g. variola virus, vaccinia virus), enteroviruses (e.g.
poliovirus, coxsackievirus), hepatitis viruses (including A, B and C), herpesviruses (e.g. Herpes simplex virus, varicelia-zoster virus, cytomegalovirus, Epstein-Barr virus), rotaviruses, Norwalk viruses, hantavirus, arenavirus, rhabdovirus (e.g. rabies virus), retroviruses (including HIV, HTLV-1 and -II), papovaviruses (e.g. papillomavirus), polyomaviruses, and picornaviruses, and the like), and bacteria (including a wide variety of pathogenic and non-pathogenic prokaryotes of interest including Bacillus; Vibrio, e.g. V. cholerae; Escherichia, e.g. Enterotoxigenic E. coli, Shigella, e.g. S.
dysenteriae; Salmonella, e.g. S. typhi; Mycobacterium e.g. M. tuberculosis, M.
leprae; Clost(dium, e.g.
C. botulinum, C. tetani, C. difficile, C.perfringens; Cornyebacterium, e.g. C.
diphtheriae; Streptococcus, S. pyogenes, S. pneumoniae; Staphylococcus, e.g. S. aureus; Haemophilus, e.g.
H. intluenzae;
Neisseria, e.g. N. meningitidis, N. gonorrhoeae; Yersinia, e.g. G. IambliaY.
pestis, Pseudomonas, e.g.
P. aeruginosa, P. putida; Chlamydia, e.g. C. trachomatis; Bordetella, e.g. B.
pertussis; Treponema, e.g. T. palladium; and the like); (2) enzymes (and other proteins), including but not limited to, enzymes used as indicators of or treatment for heart disease, including creatine kinase, lactate dehydrogenase, aspartate amino transferase, troponin T, myoglobin, fibrinogen, cholesterol, triglycerides, thrombin, tissue plasminogen activator (tPA); pancreatic disease indicators including amylase, lipase, chymotrypsin and trypsin; liver function enzymes and proteins including cholinesterase, bilirubin, and alkaline phosphotase; aidolase, prostatic acid phosphatase, terminal deoxynucleotidyl transferase, and bacterial and viral enzymes such as HIV protease; (3) hormones and cytokines (many of which serve as ligands for cellular receptors) such as erythropoietin (EPO), thrombopoietin (TPO), the interleukins (including IL-1 through IL-17), insulin, insulin-like growth factors (including IGF-1 and -2), epidermal growth factor (EGF), transforming growth factors (including TGF-a and TGF-(3), human growth hormone, transferrin, epidermal growth factor (EGF), low density lipoprotein, high density lipoprotein, leptin, VEGF, PDGF, ciliary neurotrophic factor, prolactin, adrenocorbcotropic hormone (ACTH), calcitonin, human chorionic gonadotropin, cotrisol, estradiol, follicle stimulating hormone (FSH), thyroid-stimulating hormone (TSH), leutinzing hormone (LH), progeterone, testosterone, ; and (4) other proteins (including a-fetoprotein, carcinoembryonic antigen CEA.

In addition, any of the biomolecules for which antibodies may be detected may be detected directly as well; that is, detecfion of virus or bacterial cells, therapeutic and abused drugs, etc., may be done directly.

Suitable target analytes include carbohydrates, including but not limited to, markers for breast cancer (CA15-3, CA 549, CA 27.29), mucin-like carcinoma associated antigen (MCA), ovarian cancer (CA125), pancreatic cancer (DE-PAN-2), and colorectal and pancreatic cancer (CA 19, CA 50, CA242).

The adapter sequences may be chosen as outlined above. These adapter sequences can then be added to the target analytes using a variety of techniques. In general, as described above, non-covalent attachment using binding partner pairs may be done, or covalent attachment using chemical moieties (including linkers).

Once the adapter sequences are associated with the target analyte, including target nucleic acids, the compositions are added to an array. In one embodiment a plurality of hybrid adapter sequence/target analytes are pooled prior to addition to an array. All of the methods and composifions herein are drawn to compositions and methods for detecting the presence of target analytes, particularly nucleic acids, using adapter arrays.

Advantages of using adapters include but are not limited to, for example, the ability to create universal arrays. That is, a single array is utilized with each capture probe designed to hybridize with a specific adapter. The adapters are joined to any number of target analytes, such as nucleic acids, as is described herein. Thus, the same array is used for vastly different target analytes. Furthermore, hybridization of adapters with capture probes results in non-covalent attachment of the target nucleic acid to the microsphere. As such, the target nucleic/adapter hybrid is easily removed, and the microsphere/capture probe can be re-used. In addition, the construction of kits is greatly facilitated by the use of adapters. For example, arrays or microspheres can be prepared that comprise the capture probe; the adapters can be packaged along with the microspheres for attachment to any target analyte of interest. Thus, one need only attach the adapter to the target analyte and disperse on the array for the construction of an array of target analytes.

Once made, the compositions of the invention find use in a number of applications. In a preferred embodiment, the compositions are used to probe a sample solution for the presence or absence of a target sequence, including the quanfification of the amount of target sequence present.

For SNP analysis, the ratio of different labels at a par6cular location on the array indicates the homozygosity or heterozygosity of the target sample, assuming the same concentration of each readout probe is used. Thus, for example, assuming a first readout probe comprising a first base at the readout position with a first detectable label and a second readout probe comprising a second base at the readout position with a second detectable label, equal signals (roughly 1:1 (taking into account the different signal intensities of the different labels, different hybridization efficiencies, and other reasons)) of the first and second labels indicates a heterozygote. The absence of a signal from the first label (or a ratio of approximately 0:1) indicates a homozygote of the second detection base;
the absence of a signal from the second label (or a ratio of approximately 1:0) indicates a homozygote for the first detection base. As is appreciated by those in the art, the actual ratios for any particular system are generally determined empirically. The ratios also allow for SNP
quantitafion The present invention also finds use as a methodology for the detecfion of mutations or mismatches in target nucleic acid sequences. For example, recent focus has been on the analysis of the relationship between genetic variation and phenotype by making use of polymorphic DNA
markers. Previous work utilized short tandem repeats (STRs) as polymorphic positional markers;
however, recent focus is on the use of single nucleotide polymorphisms (SNPs), which occur at an average frequency of more than 1 per kilobase in human genomic DNA. Some SNPs, parficularly those in and around coding sequences, are likely to be the direct cause of therapeutically relevant phenotypic variants. There are a number of well known polymorphisms that cause clinically important phenotypes; for example, the apoE2/3/4 variants are associated with different relative risk of Alzheimer's and other diseases (see Cordor et al., Science 261(1993). Multiplex PCR amplification of SNP loci with subsequent hybridization to oligonucleotide arrays has been shown to be an accurate and reliable method of simultaneously genotyping at least hundreds of SNPs; see Wang et al., Science, 280:1077 (1998);
see also Schafer et al., Nature Biotechnology 16:33-39 (1998). The compositions of the present invention may easily be substituted for the arrays of the prior art.

Generally, a sample containing a target analyte (whether for detection of the target analyte or screening for binding partners of the target analyte) is added to the array, under conditions suitable for binding of the target analyte to at least one of the capture probes, i.e.
generally physiological conditions. The presence or absence of the target analyte is then detected. As will be appreciated by those in the art, this may be done in a variety of ways, generally through the use of a change in an optical signal. This change can occur via many different mechanisms. A few examples include the binding of a dye-tagged analyte to the bead, the production of a dye species on or near the beads, the destruction of an existing dye species, a change in the optical signature upon analyte interaction with dye on bead, or any other optical interrogatable event.

In a preferred embodiment, the change in optical signal occurs as a result of the binding of a target analyte that is labeled, either directly or indirectly, with a detectable label, preferably an optical label such as a fluorochrome. Thus, for example, when a proteinaceous target analyte is used, it may be either directly labeled with a fluor, or indirectly, for example through the use of a labeled anfibody.
Similarly, nucleic acids are easily labeled with fluorochromes, for example during PCR amplification as is known in the art. Alternatively, upon binding of the target sequences, a hybridization indicator may be used as the label. Hybridization indicators preferenfially associate with double stranded nucleic acid, usually reversibly. Hybridization indicators include intercalators-and minor and/or major groove binding moieties. In a preferred embodiment, intercalators may be used;
since intercalation generally only occurs in the presence of double stranded nucleic acid, only in the presence of target hybridization will the label light up. Thus, upon binding of the target analyte to a capture probe, there is a new optical signal generated at that site, which then may be detected.

Alternatively, in some cases, as discussed above, the target analyte such as an enzyme generates a species that is either directly or indirectly optical detectable.

Furthermore, in some embodiments, a change in the optical signature may be the basis of the optical signal. For example, the interaction of some chemical target analytes with some fluorescent dyes on the beads may alter the optical signature, thus generating a different opfical signal.

As will be appreciated by those in the art, in some embodiments, the presence or absence of the target analyte may be done using changes in other optical or non-optical signals, including, but not limited to, surface enhanced Raman spectroscopy, surface plasmon resonance, radioactivity, etc.

The assays may be run under a variety of experimental conditions, as will be appreciated by those in the art. A variety of other reagents may be included in the screening assays.
These include reagents like salts, neutral proteins, e.g. albumin, detergents, etc which may be used to facilitate optimal protein-protein binding and/or reduce non-specific or background interactions.
Also reagents that otherwise improve the efficiency of the assay, such as protease inhibitors, nuclease inhibitors, anti-microbial agents, etc., may be used. The mixture of components may be added in any order that provides for the requisite binding. Various blocking and washing steps may be utilized as is known in the art.

50,913-4 in addition, the present invention provides kits for the reactions of the invention, comprising components of the assays as outlined herein. In addi6on, a variety of other reagents may be inciuded in the assays or the kits. These include reagents like salts, neutral proteins, e.g. albumin, detergents, etc which may be used to faciiitate optimai protein-protein binding and/or reduce non-specific or background interactions. Also reagents that otherwise improve the efficiency of the assay, such as protease inhibitors, nuclease inhibitors, anti-microbiai agents, etc., may be used. The mixture of components may be added In any order that provides for the requisite activity.

Claims (23)

CLAIMS:

1. A method of sequencing a plurality of target nucleic acids each comprising a first domain and an adjacent second domain, said second domain comprising a plurality of detection positions, said method comprising:

a) providing a plurality of hybridization complexes each comprising a target sequence and a sequencing primer that hybridizes to the first domain of said target sequence, wherein said hybridization complexes are attached to sites on an array, said array comprising at least a first substrate with a surface comprising individual sites;

b) extending each of said primers by the addition of a first nucleotide to the first detection position using a first enzyme to form an extended primer; and c) detecting the release of pyrophosphate (PPi) to determine the type of said first nucleotide added onto said primers, wherein said release of PPi is detected by secondary enzymes, said secondary enzymes being attached to sites on said array.

2. The method according to claim 1 wherein said hybridization complexes are attached to microspheres, said microspheres being associated with discrete individual sites on said surface of said substrate.

3. The method according to claim 1 or 2 wherein said substrate is an optical fiber array.

4. The method according to any one of claims 1 to 3 wherein said sequencing primers are attached to said surface.

5. The method according to any one of claims 1 to 3 wherein each of said hybridization complexes comprises said target sequence, said sequencing primer and a capture probe covalently attached to said surface.

6. The method according to any one of claims 1 to 3, wherein each of said hybridization complexes comprises said target sequence, said sequencing primer, an adapter probe and a capture probe covalently attached to said surface.

7. The method according to claim 5 or 6, wherein said microspheres comprise subpopulations of microspheres, wherein each microsphere in a subpopulation contains the same capture probe.

8. The method according to any one of claims 5 to 7 wherein said capture probe also serves as said sequencing primer.

9. The method according to claim 7 or 8 wherein said microspheres further comprise an indentifier binding ligand that will bind a decoder binding ligand.

10. The method according to any one of claims 1 to 3 further comprising:

a) extending said extended primer by the addition of a second nucleotide to the second detection position using said enzyme; and b) detecting the release of pyrophosphate (PPi) to determine the type of said second nucleotide added onto said primers.

11. The method according to any one of claims 1 to 3 comprising:

a) contacting said PPi with a second enzyme that converts said PPi into ATP; and b) detecting said ATP using a third enzyme.

12. The method according to claim 11 wherein said second enzyme is sulfurylase, and said third enzyme is luciferase.

13. The method according to any one of claims 2 to 12 wherein said microspheres are randomly distributed on said discrete sites.

14. The method according to any one of claims 1 to 13 wherein said substrate is selected from the group consisting of glass and plastic.

15. A kit for nucleic acid sequencing comprising:
a) a composition comprising:

i) a substrate with a surface comprising discrete sites, and ii) a population of microspheres distributed on said sites; wherein said microspheres comprise capture probes for hybridizing to target nucleic acid sequences;

b) an extension enzyme for enzymatically extending an oligonucleotide chain;

c) dNTPs; and d) secondary enzymes for detecting the release of pyrophosphate (PPi), said secondary enzymes being attached to microspheres.

16. The kit according to claim 15, wherein said substrate is an optical fiber array.

17. The kit according to claim 15 or 16 wherein said secondary enzymes comprise:

a) a second enzyme for the conversion of pyrophosphate (PPi) to ATP; and b) a third enzyme for the detection of ATP.

18. The kit according to claim 17 wherein said second enzyme comprises sulfurylase and said third enzyme comprises luciferase.

19. The kit according to claim 15 or 16 wherein said dNTPs are labeled.

20. The kit according to claim 19 wherein each dNTP
comprises a different label.

21. The kit according to any one of claims 15 to 20 wherein said population of microspheres comprises subpopulations of microspheres, each microspheres in each said subpopulation containing the same capture probe.

22. The kit according to any one of claims 15 to 21 wherein said discrete sites are wells, and said microspheres are randomly distributed in said wells.

23. The kit according to any one of claims 15 to 22 wherein said substrate is selected from the group consisting of glass and plastic.